LEGON JOURNAL OF THE HUMANITIES VOLUME 20 2009 Editor Gordon S.K. Adika Assistant Editor Kofi Ackah ISSN 0855-1502 ii INFORMATION FOR CONTRIBUTORS CONTRIBUTORS Legon Journal of the Humanities is published once a year by the Faculty of Arts, University of Ghana. It publishes the following types of contributions: (1) general or technical articles in the fields of research dealing with the humanities from scholars in all countries and (2) review articles on new, particularly noteworthy books. The language of publication is English. MANUSCRIPT FORM Manuscripts should be typed doubled space (including endnotes, questions and reference list) on 8.5” x 11” or A4 paper. An article should not exceed 40 pages. Three copies of each article are to be submitted to “The Editor, Legon Journal of the Humanities, University of Ghana, Faculty of Arts, P.O. Box LG 69, Legon, Ghana. ABSTRACT Each manuscript should be accompanied by an abstract of not more than 100 words. CITATIONS In the text, citation should take the following forms: (1) New evidence can be found in Yeboah (1993); (2) Brown (1979) claims that his theory marks a new era in social research; and (3) She argued for a new policy that will “promote the dignity of both men and women” (Agbo 1994: 17). Citations should be done in the text, not in the endnotes. NOTES Notes commenting or explaining points made in the text should appear after the main text, not at the bottom of the page. They should be double-spaced. SYMBOLS Conventional, current type-faces are to be used. Special symbols and diagrams should be avoided as much as possible. REFERENCES All works consulted should be listed at the end of each article under the headline iii REFERENCES. Authors should be listed alphabetically. The entries should consistently follow the examples below: Kachru, B.B. 1992. World Englishes: Approaches, issues and resources. Language Teaching. 24:1-14. Sey, K.A. 1973. Ghanaian English. London: Macmillan. Yankah, K. 1995. Speaking for the Chief: The Politics of Akan Royal Oratory. Bloomington: Indiana University Press. AUTHOR IDENTIFICATION Each article should be accompanied by a separate sheet indicating the title of the paper as well as the following information about the author; (1) full name (family name last), (2) institutional affiliation; and (3) current status (e.g. Lecturer, Senior Lecturer, Professor, etc.). For the purpose of blind reviews, the first page of each article should not bear the name of the author. EDITORIAL BOARD Kweku Osam ………………………………………………….………. Chairman Gordon S.K. Adika ………………………………………………………... Editor Kofi Ackah ……………………………..…………………………… Asst. Editor Helen Yitah …………………………………………..……………….… Member George Ossom-Batsa ………………………………………………...… Member Augustine H. Asaah ………………………………….…………………. Member INTERNATIONAL ADVISORY BOARD David Owusu-Ansah James Madison University, U.S.A. Emmanuel K. Akyeampong Harvard University, U.S.A. Laud Ato Quayson University of Toronto, Canada EDITORIAL CORRESPONDENCE Editorial correspondence should be addressed to: The Editor, Legon Journal of the Humanities, Faculty of Arts, University of Ghana, P.O. Box LG 69, Legon, Ghana. E-mail: adika@ug.edu.gh or editorljh@ug.edu.gh iv SUBSCRIPTION INFORMATION Business correspondence should be addressed to: The Manager, University Bookshop, P.O. Box LG 1, Legon, Ghana. E-mail:bookshop@ug.edu.gh THE SUBSCRIPTION RATES ARE AS FOLLOWS: Local ………………………………………………………………… GH¢ 4 Overseas Individuals ………………………………………….... US$ 20 (Airmail)a. Libraries & Institutions …………………………...… US$ 35 (Airmail)b. vCONTENTS Relationship between Religion and Science Kwame Gyekye .............………………………………… 1 The “Paranormal”: African Philosophy Questions Science Martin Ajei ……………..……………………………….. 23 Subjects, Agents, or Collectives? The Discourse of Youth and Philosophy Ibrahim Bello-Kano ………………..…………………… 45 Globalization of an African Language: Truth or Fiction? Josephine Dzahene-Quarshie …………………………... 69 Corruption and the Body Politic in Post-Colonial Ghana: A Re-reading of Amu Djoleto’s Money Galore in the era of ‘Zero tolerance for Corruption’ in Ghana Mawuli Adjei …………………….……………………… 89 Speech Surrogates of Africa: A Study of the Fante Mmensuon Adwoa Arhine ………………………….……………….. 105 The Phonology-Syntax Interface in Avikam Firmin Ahoua …………………………………………… 123 Double Plurality in Codeswitching Evershed Kwasi Amuzu ………………………………... 151 Comparing Demonstratives in Kwa Nana Aba Appiah Amfo and Stella Boateng Norgbe …. 181 1RELATIONSHIP BETWEEN RELIGION AND SCIENCE: AN OVERVIEW Kwame Gyekye1 Abstract This paper presents an overview of the relationship between religion and science. It points out that historically religion preceded science, as the limitations of human intelligence in a bizarre world led man very early to postulate a being considered ultimate, supreme and worthy of human obeisance and worship. Like religion and philosophy, science began in wonder: to explore the wonders of nature – of the physical world. Religion and science are related in that both of them have perspectives on cosmic reality, even though there are several differences in their interpretations of reality. It is the different interpretations as well as their methods at arriving at their truths and conclusions that eventuated in conflicts, conflicts that actually came to the fore with the emergence of experimental science in and after the seventeenth century of our era and led to the condemnation by the Catholic Church of Galileo, the acknowledged founder of modern science. Scientific theories such as the evolution theory, quantum physics, and some theories of neuroscience presented challenges to religious doctrines of creation, cosmic order and intelligibility, divine sovereignty, and human nature. However, there are areas of integration, such as natural theology and design, order and regularity of nature, that provide evidence of the existence of God – evidence that is supported by most scientists. The paper concludes that religion and science are different languages that ultimately express the same reality or at least present complementary accounts of reality, and that, given the wonders and mysteries of the created universe and the limitations of human intelligence, religion and science will continue to be bedfellows in the twenty first century and beyond. 1. Introduction Before science, religion was. This was because the conviction by man of his own limitations in a bizarre world that is full of incomprehensible events or phenomena led him to postulate a being and a cause considered unlimited and ultimate. Even though it is possible for science to throw doubts in the minds of people about the status of religious belief and, thus, lessen the enthusiasm for religion, there is no credible evidence that religious belief or experience will disappear or lose its influence on human beings in the wake of the emergence of science. What seems 1 Professor Kwame Gyekye (FGA) is a Professor of Philosophy in the Department of Philosophy, University of Ghana, Legon. 2 Legon Journal of the HUMANITIES Volume 20 (2009) to have happened, however, is that both the religious and scientific enterprises have come to be embraced and are influencing human attitudes to nature or the universe, though not without mutual suspicions or even conflicts, as will be pointed out in detail in this paper. I define religion as the awareness of the existence of some ultimate supreme being held as the origin and sustainer of this universe and the establishment of constant, generally worshipful, ties with this being. Thus defined, a religion would evolve and maintain a system of beliefs about the totality of human experience. I define science simply as nature study – as the intellectual enterprise concerned with investigations into natural phenomena, into the structures of the physical world. The two enterprises – religion and science – are related in that they both have perspectives on reality, even though their interpretations of reality differ in several ways. The relationship is based also on the fact that they both affect our attitudes to this complex world differently. In this paper, I explore the relationship between religion and science. I must state at once that I am not a scientist. I am a philosopher who shows some interest in science. My exploration is based on the statements, views and arguments of scientists, as much of them as I understand. I will confine myself, not to the details of the scientific experiments that gave rise to scientific statements or conclusions, but to the logic of scientific statements. In philosophy of science, philosophers seek to clarify or analyze concepts, statements and arguments of science. 2. Relationship between Religion and Science before the 17th Century A.D. For many, many centuries the status of religious belief was acknowledged and was unquestioned. Not that observations and inquiries into nature had been lacking in the centuries preceding the seventeenth century of our era, which is often regarded as the starting point of modern science; for, ancient thinkers, particularly the Pre- Socratics, Plato, Aristotle, and Epicurus did make extensive observations about nature or the physical world. Atomism, which is a theory of the origin of the physical world, was in the Western world developed by the philosophers Leucippus (fl. 440 BC) and his disciple Democritus (460-370 BC) and further developed by Epicurus (341-270 BC) and in the Roman world by the poet and philosopher Lucretius in the first century BC. Atomism spread to the medieval Western world largely during the Renaissance, with the work of philosophers such as Nicholas de Cusa (1401-1464), the astronomer and natural philosopher Galileo Galilei (1564- 1642), who found atomism to be consistent with his experiments in physics, and others. Let me mention that the atomic doctrine of Democritus contains the Kwame Gyekye 3 statement: “Nothing can be created out of nothing, nor can it be destroyed and returned to nothing.”1 This statement could have been troubling to religious belief, for it obviously flew in the face of the doctrine of creatio ex nihilo (‘creation out of nothing’), a basic doctrine of Christianity and other orthodox religions. Aristotle’s work titled Physics was widely read and discussed during the medieval period; it became very influential. Among Aristotle’s assertions are that the world was uncreated (ungenerated: Greek ageneton), indestructible (Greek: aphtharton), and eternal (Greek: aion), that time and motion are incorruptible: they have no beginning and end. The philosophical ideas of Aristotle reached the Medieval Christian world through translations from the Arabic. Even though they were challenged, they do not, for some curious reasons, seem to have received the official condemnation of the theologians as did the ideas of Galileo or Charles Darwin in subsequent centuries. The theory of the medieval astronomer Copernicus, that the sun rather than the earth was the centre of the planetary system, appeared to conflict with traditional Christian religious ideas, just as it was in defiance of the astronomy of Ptolemy. Despite all this, the authority and influence of religion remained largely unscathed in those centuries, i.e., before the 17th Century AD. I think one reason that led to this was that science, then, was not an autonomous intellectual enterprise based on and guided by systematic experimentation. Even though observation was not absent from Aristotelian and medieval science, nevertheless, their theories could not be tested by further experiment. Thus, not having been able to wean itself from its Greek antecedents, medieval science became a branch of philosophy, to all intents and purposes. Lacking a basis in experimentation, it could not discover or arrive at far-reaching significant scientific ideas that could be considered subversive of orthodox religious doctrine. Without a basis in experimentation, medieval science, like its Greek precursor, was essentially deductive rather than inductive; whereas in its method science is essentially or primarily inductive, notwithstanding the deductive features of the scientific method that are manifested particularly in the application of mathematics to the physical sciences. The overall consequence of the interaction between science and religion in the period before the 17th century was a synthesis between them: science was given accommodation and embrace in the house of God, so to speak. The conflicts that characterized the relationship between religion and science in and after the 17th Century did not occur in the period before then. 4 Legon Journal of the HUMANITIES Volume 20 (2009) 3. The Emergence of Modern Science The emergence of modern experimental science later in the 17th Century was to change the previously symbiotic relationship between religion and science; it brought about hostility and conflict in the relationship. The new science of the 17th Century and beyond showed itself as a restless and belligerent spirit, unprepared to make overtures with religion or any kind of metaphysics that stood in its way of arriving at what it saw as the truths about the universe. Galileo, regarded as the father of modern science because of his emphasis on and commitment to experimentation as the new scientific methodology, together with other investigators of the seventeenth century, set science free from the restraining influences of the Church and made it an independent sphere of human thought. Galileo placed himself on a collision course with the Church when he advocated the new Copernican theory which maintained that the earth and the planets revolve around the sun, rather than the accepted Ptolemaic theory in which the sun and the planets revolve around the earth. The Ptolemaic theory was said to be in harmony with scriptural passages that implied that the earth is the centre of the cosmos: the sun and the planets revolve around the earth because the earth is the centre of the cosmos. Galileo was, thus, seen as challenging the authority of the Church and was condemned after he was tried before a panel of cardinals in 1633. He was condemned for violating an injunction sent to him in 1616 requiring him “not to hold, teach, or defend in any way whatsoever that the earth moves.”2 We learn, however, that “throughout the controversy, and until his death, Galileo remained a religious man.”3 Galileo gave up his scientific belief and spent the rest of his life under house arrest. He was forbidden to publish his work titled Discourses on Two New Sciences but managed to get the work published in Protestant Holland in 1638, four years before his death in 1642. The Dialogues of Galileo remained on the Index of Prohibited Books until 1822. In 1984 Pope John Paul 11 appointed a commission to reexamine those events. The commission admitted that “church officials had erred in condemning Galileo.” 4 In 1992 after reviewing the commission’s findings, the pope said that there are “two realms of knowledge” and that the failure to distinguish them had led theologians “to transpose into the realm of the doctrine of the faith a question that in fact pertained to scientific investigation.” 5 The pope was, thus, asserting the Independence thesis on the relationship between religion and science. I will explain the thesis shortly. Even though Galileo’s trial appeared to be episodic and exceptional, it presaged what was ahead in the relationship between religion and science in the decades and centuries to come. There followed long periods of interminable Kwame Gyekye 5 hostilities, wrangles, disputes, controversies, and disagreements over the relationship between religion and science that have continued to this day and are likely to go beyond this century. It would be correct to say that conflicts or sharp disagreements between religion and science arise basically because both of them are concerned with the interpretation of reality, with the search for that which is ultimately or absolutely real. The conflict is, thus, grounded on the different perspectives on reality. But the conflict derives also from their methods in arriving at their truths or conclusions. Science requires explanations that can be generalized, facts that are disciplined by experimentation, and experiments that are repeatable and verifiable elsewhere: thus, scientific methods are objective. Science is concerned about causal relations between empirical events. Religion, on the other hand, is subjective and is concerned about meaning and purpose of life. The question arises as to which of them, if either, offers a more credible or satisfying perspective on reality. Ian Barbour, a professor of physics and religion, has identified four ways in which science and religion may be said to interact.6 These ways are: Conflict, Independence, Dialogue, and Integration. Not being a scientist and, thus, not deeply knowledgeable about the really technical or professional details and nuances of scientific arguments, I would avoid such nuanced arguments, even though I hope to say much that will enable us to appreciate the issues involved in the relationship between religion and science as well as the attempts to deal with them. Some scientific theories, such as the evolutionary theories, raise philosophical questions and can be explored from the philosophical point of view. The Conflict view of the relationship between religion and science simply means that religious doctrine and scientific perspective are incompatible and so cannot in any way be reconciled. It means, therefore, that one cannot logically accept the doctrines or positions of both, for the position of one excludes the other; a person can only choose one or the other, not both. I will in due course provide several instances where religious doctrines are alleged to be in conflict with scientific positions. According to the Independence thesis, conflicts between religion and science need not arise because religion and science refer to different aspects of reality, employ distinctive methods, serving totally different functions in human life as reflected in their different languages; their objects also are different – science dealing with questions about natural phenomena and, thus, about objective facts, while religion deals with ultimate meaning and purpose and recommends allegiance to particular moral principles and a way of life oriented to the supernatural or the hereafter. Thus, the two are parallel conceptual systems, and because they do not 6 Legon Journal of the HUMANITIES Volume 20 (2009) cross each other’s path, there should be no conflict between them, according to the Independence thesis. They are separate and mutually exclusive realms of human thought and experience, each realm having its own distinctive questions, rules and criteria of judgment. The Independence thesis, thus, tries to steer clear of the Scylla of insistent and unyielding theism and the Charybdis of restless and self- assured scientific enterprise. On the surface, the Independence view may appear attractive particularly to the non-scientist person with religious faith, but not to the scientist with some religious faith. However, the Independence view of the relationship between religion and science prevents any constructive interaction between them, such as the influence one of them could have on the other, the inspiration one of them could derive from the other, and the mutually relevant and valuable questions one of them could raise for the other’s attention, and so on. The assumption of the Dialogue view is that while the differences between the methods of religion and science may be different, there may be similarities which must be acknowledged. Science is interested in the question as to why the universe is orderly and intelligible, for instance, but it is not in a position to answer the question satisfactorily. In this matter, through a dialogue between religion and science, religion might be able to provide some answer that may be of interest to the scientist. Similarly, dialogue may arise when one field employs for its own purposes analogous concepts analyzed in the other field. Thus, analogous concepts in science are used by religion to talk about God’s relation to the world. The valuable thing about Dialogue view is that it emphasizes similarities between religion and science such as can be discovered, whereas the Independence view merely emphasizes the differences between them. The Integration view argues that religion and science can contribute to the development of a metaphysic that has the potential of satisfying the demands and goals of both. This view, however, requires that some of the traditional religious beliefs or doctrines should be reformulated in the light of scientific discoveries or theories. 4. Natural Theology and Design One example of the Integration view is natural theology, which infers the existence of God from the evidence of design in nature, evidence that is supported or confirmed by most scientists. The founders of modern science, says Barbour, “frequently expressed admiration for the harmonious coordination of nature, which they saw as God’s handiwork. Newton said that the eye could not have been contrived without skill in optics, and Robert Boyle extolled the evidences of Kwame Gyekye 7 benevolent design throughout the natural order.”7 Natural religion or theology has empirical basis for, like science, it draws on, or takes off from, natural phenomena and man’s reflections on these phenomena. Natural theology, which also leads to the postulation of the existence of God, results from man’s application of reason to the characteristics of nature as an object. Design or the argument based on design is discovered through man’s rational enterprise. Traditional African religion, not being a revealed religion, is, I have argued elsewhere, a natural religion.8 Natural theology, which derives from design, can support revealed theology. It is a conception of design in nature – a design that derives from an act of God – that grounds the orderliness and intelligibility of nature. It would be correct to say that much, if not the entire, enterprise of science operates from the orderliness of nature and the regularity that characterizes natural events or phenomena. Orderliness and regularity, expressed or manifested in the laws of nature, are fundamental presuppositions of science. It is the orderliness of nature and the regularity that follows from it that make predictability in science possible. But design also implies that the world in which science functions is a determinist world, a world in which events happen according to the laws of nature. Thus, knowledge of all antecedent conditions and laws of nature would make it possible to predict the entire future. This means that determinism excludes chance. The source of determinism in nature must be a Great Intelligence – the God of the theistic religions, a conscious being. The seventeenth century physics of Sir Isaac Newton upheld the determinism of the world. Newton maintained the idea of a universe rigidly determined by natural laws fixed by an intelligent creator, God. Newton believed that God “keeps the stars from collapsing under gravitational attraction and intervenes periodically to correct planetary perturbations in the solar system.”9 Thus, for Newton and his followers, God not only designed the laws of nature but sustains them continually. This act of sustenance is an expression of God’s purpose and sovereignty. In a determinist world there is no place for chance or contingency; determinism and chance are incompatible. Contingency subverts regularity, orderliness and determinism. Albert Einstein, undoubtedly the greatest of the twentieth-century scientists, strongly believed in the order and predictability of the universe, which, like Isaac Newton, he maintained was a determinist universe. He considered contingency a threat to belief in the rationality or intelligibility of the world, a feature of the world which he thought is central in science. Science merely assumes the intelligibility of the world; but this feature of the world is an aspect of the creative act of God, the creator. Einstein noted: “A conviction, akin 8 Legon Journal of the HUMANITIES Volume 20 (2009) to religious feeling, of the rationality or intelligibility of the world lies behind all scientific work of a high order.”10 He expressed “a deep faith in the rationality of the world.”11 And, in searching for unified laws in cosmology, the physicist James Trefil writes: But who created those laws? ...Who made the laws of logic? ... No matter how far the boundaries are pushed back, there will always be room both for religious faith and a religious interpretation of the physical world. For myself, I feel much comfortable with the concept of a God who is clever enough to devise the laws of physics that make the existence of our marvelous universe inevitable.12 5. Quantum Physics and Religion Physics, the study of the basic structures and processes of change in matter, was undoubtedly the first science that was systematic. Its Greek root, phusis, means ‘nature’, i.e., the material or sensible world of our everyday experience. Thus, the earliest Greek thinkers who paid attention to inquiries about nature, such as the atomists, were called physikoi (‘physicists’) or physiologoi (‘physical speculators’, ‘speculators of nature’, ‘philosophers of nature’). It would be correct, in my view, to say that, among the sciences, physics has exerted the greatest influence on philosophy and theology. Classical physics upheld the deterministic character of the world and, so, did not present any challenge to religious beliefs. However, deism, the belief that God created the universe and left it to run by itself, thus restricting God’s role to that of a clockmaker, was developed in the eighteenth century.13 In implying that God does not intervene or operate continuously in the affairs of the world after having created it, deism proposed a doctrine that was at odds with fundamental religious beliefs. Classical physics affirmed the certainties of prediction. By contrast, quantum physics, developed in the 1920s and was about atomic and subatomic phenomena, maintained that there were inherent uncertainties in the predictions of events. Quantum physics was a rejection of determinism on one hand and an acknowledgement of the openness of the future and its consequent indeterminacy on the other hand. (I must confess that I do not understand the intricacies and nuances of the quantum theory of physics, just a little enough to enable me to relate some aspects of it to religious beliefs.)14 It is the indeterminacy in nature itself that gives rise to the uncertainties in the predictions made by quantum theory. The consequence of the rejection by quantum theory of determinism is that quantum events occur by chance. The notion of chance in quantum phenomena challenges ideas of divine purpose and Kwame Gyekye 9 sovereignty or control. Chance, like indeterminacy, limits the power of God. It is known, however, that a minority of physicists, including Einstein and Max Planck, have stuck to the deterministic theory, maintaining that the uncertainties of quantum theory are to be attributed to temporary human ignorance and that someday appropriate physical laws will be found that will make accurate predictions possible. Einstein wrote: “The great initial success of quantum theory cannot convert me to believe in that fundamental game of dice…. I am absolutely convinced that one will eventually arrive at a theory in which the objects connected by laws are not probabilities but conceived facts.”15 And, in a famous statement that rejects the element of chance, he said: “God does not play dice.”16 To this famous statement of Einstein’s Niels Bohr, a Danish physicist considered the founder of modern atomic physics, responded, “Nor is it our business to prescribe to God how He should run the world.”17 Bohr’s point, I think, is that we should remain agnostic with regard to God’s attitude to the world: we do not know, so would Bohr say, whether God plays dice with the world or not. Einstein expressed his confidence in the order and predictability of the universe, which he thought would be damaged by any element of chance that derives from indeterminism – from quantum physics. We must note, however, that a number of physicists deny that uncertainty is the result of temporary ignorance but that it is a fundamental limitation that hinders the achievement of exact knowledge of the atomic world. And, I have learned that the French physicist Alan Aspect and his group have “in a series of elegant experiments confirmed the correctness of the quantum mechanical prediction.”18 The logic of the arguments of anti-Einstein physicists (who may be referred to as ‘the indeterminists’), then, is that it is possible to reject determinism, install chance, and yet believe in scientific predictions. What I still find mind- boggling, in the wake of the quantum rejection of determinism, is the possibility of scientific predictions being based on chance, on randomness, on irregularity! Quantum theory seems to present a paradox. It would be correct to say, however, that even though quantum physics’ notions of indetermination and chance must originally have been perceived as in conflict with religious doctrine, nevertheless, the logic of quantum physics appears, at least in part, compatible with religious belief. First, the notion of the openness of the future, which is a logical feature of indetermination, suggests that the human mind or human knowledge is limited and so cannot see far into the future. Beliefs in the limitations of human knowledge are compatible with Biblical ideas. Second, some theologians have argued that the notion of quantum indeterminacy allows room for divine intervention and action: thus, that God determines the 10 Legon Journal of the HUMANITIES Volume 20 (2009) indeterminacies left open by the laws of quantum physics.19 If, indeed, God controls or determines all indeterminacies, then it would follow that the traditional idea of predestination could be preserved. Chance would, consequently, have been eliminated. What appears to us human beings as chance must be held as really determined by God. In connection with this, let me say that for Akan thinkers, as for Aristotle, a chance event as such would, in fact, be an event whose cause is unknown, not one lacking a cause and occurring randomly.20 The logical implications of quantum indeterminacy chime in with the religious doctrine that God controls all the events that appear to human beings as chance events. As part of the development of the quantum theory, Niels Bohr proposed the principle of complementarity, which asserts that there can be two complementary descriptions of the same reality and that for a full account and comprehension of reality both perspectives are, or would be, needed. I do not claim to comprehend the details of the principle as applied to wave picture and particle picture. But I believe that the principle can be extended and applied to the relation between religion and science, compelling us to recognize that reality is a complex phenomenon that can be grasped from different approaches, which taken together help us to understand man’s holistic experience in the world. Religion and science are different languages that ultimately express the same reality. 6. Evolution and Creation One of the central pillars of the orthodox religions is the doctrine that God created the world. And, according to the Genesis story, God created the world in seven days. The doctrine of creation has been challenged by the theory of evolution. Because the evolutionary change takes place over a very long period of time, it conflicts with the seven days of creation. The theory of evolution, which challenges the idea of creation, also challenges other religious doctrines related to creation. For instance, it eliminates design; it affirms a world without design and, thus, a world without purpose. Thus, it was not surprising that, when Charles Darwin proposed the evolutionary theory in his famous work titled On the Origin of Species (1859), he was met with vehement opposition from theologians; the latter of course maintained that evolution and religious beliefs were incompatible. It must be noted, however, that Darwin believed that God designed the whole evolutionary process but not the detailed structure of particular organisms, that the laws of evolution, not individual species, were the product of design. In his own words: “I am inclined to look at everything as resulting from designed laws, with the details, whether good or bad, left to the working out of what we may call chance…. Kwame Gyekye 11 I cannot think that the world as we see it is the result of chance; yet I cannot look at each separate thing as the result of Design.”21 Darwin’s position on design vis-à- vis chance, as presented in this quote, is not very transparent; in fact there is some confusion here. It is not clear whether he fully supports design or fully supports chance. But can both be really jointly supported? Not really. Perhaps what Darwin means is that God designed the entire evolutionary process by creating the basic laws of evolution and that inherent in the designed world were potentialities that would be actualized in due course, the actualization occurring by chance, without God having anything to do with it. The implications here fly in the face of religious doctrine. First, if my interpretation of the quotation from Darwin is correct, it makes God’s act of creation tentative and incomplete, whereas the story of the Genesis says that ‘God saw everything that he had made, and, behold, it was very good’ (Genesis 1: 31). Second, it suggests that after creating the so-called evolutionary laws God left the theatre of human affairs – the affairs of the physical world, without further divine intervention in the affairs of that world: a position that is inconsistent with the Biblical view of God’s sovereignty, his unrelenting love for man and his unflagging interest in man’s aspirations and well-being. Third, it makes the wrong suggestion that the perfect God would leave anything to chance. Evolution merely denies that our complex world was created by God; but it cannot deny that the simplest particles from which the complex world evolved were created. The question, “who created the original infinitesimal particles from which the complex world emerged?” would always be asked. To this question the response of evolutionary science in terms of chance would not be satisfactory, for it would make every event or phenomenon that occurs in the world a chance event, a position that would make the scientific enterprise itself well-nigh impossible. A former professor of physics at the University of Ghana who had been an agnostic for several years later admitted that the foundational particles out of which this complex world would have emerged could not have come about by chance but could only have been brought into being by a Great Intelligence and that, if this Great Intelligence was that identified with God, then he would believe in God.22 In short, the evolutionary theory presented several challenges to the Christian religion. Among these are: a challenge to the Biblical doctrine that God directly created the world and did so in seven days; a challenge to the doctrine that God designed the world; a challenge to the religious belief that God directly created all humanity and endowed man with a nature; and a challenge to the status of ethical values. As regards the status of ethical values, evolutionary theory would lead to the relativity or subjectivity of ethical values and would, consequently, 12 Legon Journal of the HUMANITIES Volume 20 (2009) reject the objectivity of values, for on the evolutionary theory values would be in a flux. Yet, the notion of the objectivity of values is very essential for the stability and smooth running of the human society. Religious belief, on the other hand, would generally affirm the objectivity and universality of ethical values. 7. Evolution and Human Nature According to religious doctrine, God created man directly and “in His own image” and endowed him with intelligence and moral will to be able to respond to the demands of righteousness and justice. Man is, thus, not the result of an evolutionary process, but of a creative act. The Creator must have determined certain essential or intrinsic characteristics about man. In this way, God endowed man with a nature: human nature or human essence, with its complex and ramifying implications for human behavior. Evolutionary science, however, being essentially atheistic, denies the reality of human nature, for there would be no God to have a conception of it; that is, there would be no God to have fixed or determined it. Charles Darwin’s view, already referred to, that the whole process of evolution was designed by God implied that the human soul, among other entities, was not directly or immediately created by God, a position that was in conflict with a crucial religious view of human nature. 8. Neuroscience and Human Nature The Christian religion, perhaps like other religions, maintains the body/soul dualism, that is, that a human being consists of two distinct entities or substances – body and soul (mind). They are distinct in that their natures are essentially different: while the body is a material, spatial, and mortal substance, the soul is an immaterial, non-spatial, and immortal substance. The soul is often identified with the self. (The terms mind and consciousness are used by dualist metaphysicians as equivalent to the soul or self.) The Christian dualist conception of human nature, which is anchored in the Bible, has been rejected by some neuroscientists (i.e., scientists who investigate the human brain) and by materialist philosophers influenced by the investigations of neuroscience. These brain scientists and materialist philosophers reject the body/soul or mind/body or brain/mind dualism by rejecting the nonphysical attributions of the soul or mind or self and reducing mental states or mental events to brain states or brain events. They, thus, identify mental states with brain states. In their terms, then, statements about the mind are to be translated as statements about the brain or the central nervous system, which is a physical system. Thus, for some neuroscientists and scientific materialists matter (or, the physical substance) is the fundamental reality. Mind or self is not held by them Kwame Gyekye 13 as real, certainly not as primarily real. And, consciousness may even be part of matter. But such a view is not convincing, for science cannot really account for the presence of conscious beings in the world. The reason is that no amount of subatomic particles will give rise to consciousness. Something outside the complex of physical laws is required to explain the rise of consciousness. However, the researches, explanations and arguments of some other neuroscientists have led to the affirmation of the doctrine of the body/soul or mind/brain dualism. This doctrine has a long history behind it, as it goes back to Socrates, Plato, Saint Augustine, the Church Fathers, Saint Thomas Aquinas, Rene Descartes, and many other contemporary scientists and philosophers. In their recent most detailed book titled The Self and Its Brain: An Argument for Interactionism23, the neurophysiologist Sir John Eccles (of Cambridge University, who was awarded the Nobel Prize for Neurophysiology in 1962) and the famous philosopher of science Sir Karl Popper (of London University) deploy detailed but complex explanations and arguments to defend dualism and the interaction of consciousness (i.e., soul) and the brain. It is stated in the book that Eccles, the brain scientist, is “a believer in God and the supernatural”, while Popper is “an agnostic”. Thus, in Eccles, as in many other scientists, science and religion meet. Notwithstanding what has been said in the foregoing paragraph, Barbour thinks that “most scientists today do not accept either a body/soul or a brain/ mind dualism.”24 Even so, it would be correct to assert that scientists who are Christians or hold some religious faith would affirm the body/soul dualism as an aspect of their faith, for it is the religious or metaphysical doctrine of dualism that constitutes the entire basis for beliefs in personal survival following the demise of the body. 9. The Conception of Man as a Unity A conception of man as a unity derives from Aristotle’s discussion of the soul in his De Anima. But Aristotle realizes that a conception of man as a unity of soul and body subverts the notion of personal survival after death, for, if the soul and the body are united, the disintegration of the soul concomitantly follows the disintegration of the body at death. This makes the soul a mortal – not immortal – substance. Realizing that the conclusion was fundamentally in conflict with the conceptions of the soul held in Greek thought long before him, Aristotle recoils from the idea of the unity of man. However, some contemporary scholars, according to Barbour, interpret the Bible as holding “a view of man as holistic, not dualistic.”25 He says that “many theologians today have tried to reclaim the biblical view of the self as a unified activity of thinking, feeling, willing, and acting.” 26 Thus, it is claimed that 14 Legon Journal of the HUMANITIES Volume 20 (2009) the Bible holds that man is a unity – a unity of body and soul. I doubt very much that the unity of man can really be said to be a perspective of the Bible in the light of the logical implications of this notion. One such implication, as already alluded to, is that it destroys the Biblical doctrine of the immortality of the soul and life after death. The contemporary notion of the unity of man is most probably influenced by neuroscience – the science of the brain – which generally sees the brain as that which is primarily real and to which, for some neuroscientists, mental phenomena are (to be) reduced. At this juncture, I would like to say something about the relationship between religion and science in Islam based on my knowledge of Arabic philosophy. It would be correct to say, I think, that in Islam the relationship between culture – of which science is a part – and religion has historically been an integrative and cohesive relationship; the two are intertwined, not separable. In consequence of this relationship, conflicts and disagreements between scientists and theologians do not seem to have arisen in the history of Islam, certainly not the type of interminable conflicts one can read of in Christendom. There does not appear to be much of a debate arising out of possible conflicts between science and doctrines espoused in the Qur’an (the Islamic religious book). Thus, there were no real endless, inter-generational internal conflicts as such. Conflicts that may have preoccupied the attention of Muslim theologians were scientific statements made by Western or Christian scientists that posed challenges to the doctrines of Islam. Thus, such possible challenges to Islamic religious belief were externally – not internally – induced. And attempts were made by Muslim theologians to denigrate such challenges and stem any damage they might do to Islamic belief. One example was the elaborate and complex arguments deployed by the Muslim philosophical theologian al-Ghazali (d. AD 1111) to refute some of the scientific or physical ideas of Aristotle and his adherents that had found their way into the Islamic intellectual culture through the translation of Aristotle’s works into Arabic in and after the tenth century AD. Such ideas include: the world as uncreated; the eternity and indestructibility of the world; the indestructibility (eternity) of time and motion – for they have no beginning or end; the notion of the necessary causal connection between natural events – a notion that subverts the doctrine not only of divine intervention in mundane matters but also of the existence of miracles which interrupt the usual course of nature; and other Aristotelian ideas about nature (Greek:phusis). Al-Ghazali refuted these ideas about nature in his famous work titled Tahafut al-Falasifa (translated as The Incoherence of the Philosophers). This work, which was translated into Latin and given the title of Destructio Philosophorum (‘The Destruction of the Philosophers’), was in Kwame Gyekye 15 the course of the twelfth century in turn refuted by another Muslim philosopher, Averroes (Ibn Rushd, d. 1198), in a work titled Destructio Destructionis (‘The Destruction of the Destruction’). (Averroes was a contemporary of St. Thomas Aquinas). However, Averroes’s refutation of Al-Ghazali and, thus, his defence of the physical ideas of Aristotle, did not receive much philosophical or theological attention in the Muslim world. It was Al-Ghazali’s refutation that gained theological currency and ascendancy in the subsequent decades. Now, the reason why the scientific or physical ideas of Aristotle did not attract continuous debate or discourse among Muslim scholars was because in the Islamic religious world the Word of Allah as contained in the Qur’an was held supreme and overriding, taking precedence over all other sources of knowledge: it was, thus, not subject to questioning and debate. Thus, any simmering controversy about ideas or theories of the natural world was submerged under the waves of the relentless pursuit of doctrinal orthodoxy. In this way, internally-induced conflicts that affected religious belief could not – and did not – arise within the ambience of Islamic religious doctrine. Moreover, free thinking that often results in subverting tradition and religious orthodoxy does not appear to have been a permanent and outstanding feature of intellectual life in Islam. 10. Summary At this point, I would like to summarize the issues so far and then draw some conclusions on the relationship between religion and science: Both religion and science are concerned about our understanding and i. interpretation of reality, even though their interpretations generally differ. Many of the assertions and arguments of the scientists and philosophers of ii. nature before the 17th Century of our era conflicted with religious beliefs, but they were not met with the kind of the theological belligerency that descended on the assertions and arguments of subsequent, i.e., modern scientists. The relationship between religion and science in those times was symbiotic. The emergence of modern experimental science, with its far-reaching and iii. significant discoveries and conclusions, disrupted the hitherto somewhat cozy relationship between religion and science. Natural theology proceeds from human reflections on their experiences of iv. the natural world, reflections that led them to a conviction of the existence 16 Legon Journal of the HUMANITIES Volume 20 (2009) of a supreme being that created the world; thus, natural religion, like science, has an empirical and rational foundation. While classical physics affirmed determinism and the certainties of v. prediction, quantum physics rejected determinism and rather affirmed indeterminacy in nature, the notion of chance, and the uncertainties of prediction. The notion of chance in quantum physics does violence to the concept of divine purpose and control. Quantum physics subverted the order and regularity of the universe upon which the scientific enterprise itself depends and operates. However, the logic of quantum indeterminacy casts doubts on the seriousness of its consequences for divine control and intervention. The Biblical doctrine of creation has been challenged by the theory of vi. evolution that maintains that the world was not directly created by God but evolved gradually from infinitesimal particles. Evolutionary theory eliminates design and affirms chance and randomness. It challenges Christian doctrine in several ways and rejects the objectivity of ethical values. But the evolutionary theory itself bristles with problems. Evolutionary science rejects the notion of human nature fixed by God vii. beforehand. This means that the human soul was not directly created by God, a position that is antithetical to religious belief. The dualist – i.e., soul/body – conception of human nature, affirmed by viii. all religions, is rejected by some neuroscientists (or, brain scientists) who maintain that mental states or events are brain states or events. But some other neuroscientists not only accept the soul/body dualism but defend the interaction of consciousness (soul) and brain. Aristotle had argued for a conception of man as a unity, but later had to ix. abandon it because of its consequences on the notion of personal survival at death. The claim by a number of theologians that the Bible maintains the unity of body and soul cannot be defended in view of its logical implications for the immortality of the soul. In contemporary times, the idea derives from, or is influenced by, neuroscience. Kwame Gyekye 17 11. Conclusion Let me try, in conclusion, to articulate or bring into focus the nature of the relationship between religion and science. Several attempts have been proposed that allow for interaction rather than conflict between religion and science. The interaction stems basically – and ultimately – from the fact that both religion and science are concerned fundamentally about reality. For this reason, the immediate assumption is that there must be areas of belief and goal that overlap. For this reason, religion and science must be seen as presenting complementary accounts of reality. Before the emergence of science, religion had established certain notions or doctrines about the physical world: these, or at least some of them, constituted a challenge to science and set scientists to explore them. Thus, religion may, in some way, be said to have been an important factor in the rise of science. Accordingly, Albert Einstein stated in 1948: While it is true that scientific results are entirely independent from religious or moral considerations, those individuals to whom we owe the great creative achievements of science were all of them imbued with the truly religious conviction that this universe of ours is something perfect and susceptible to the rational striving for knowledge.27 Earlier, in an article published in New York Times Magazine in1930, Einstein wrote: The cosmic religious feeling is the strongest and noblest motive for scientific research.28 And, in a Symposium on Science, Philosophy and Religion (1941), Einstein said: Science can only be created by those who are thoroughly imbued with aspirations toward truth and understanding. This source of feeling, however, springs from the sphere of religion.29 The reason for the order and intelligibility of the cosmos (universe) cannot be answered within science itself. Science will have to depend on religious belief in asserting the origin of cosmic order and intelligibility. In the phenomenon of natural theology there is a common ground between religion and science, as both of them depend on empirical experience for their rational arguments and conclusions. 18 Legon Journal of the HUMANITIES Volume 20 (2009) Scientific knowledge can be utilized to tease out religious doctrines, i.e., to clarify and help remove obscure points in religious doctrine, as in the complementarity principle of Niels Bohr. That the world is based on design by a supreme intelligence – God – is a proposition accepted largely by both science and religion, though with necessary adjustments. Nature as a law-abiding machine expressed in fixed natural laws constitutes the basis of the order and regularity of the universe, which make scientific prediction possible and generally accurate. God cannot be swept away by the natural laws he established. God might be conceived to act in ways consistent with scientific theories, even though it is possible for Him, by virtue of divine sovereignty and omnipotence, to act in ways that disestablish the established laws of nature Even though quantum physics appears to reject the determinism of natural laws and, thus, affirms indeterminacy and chance in nature as well as the uncertainties of scientific predictions, nevertheless, it appears – from the positions of Einstein, Max Planck, Laplace, and others – that the assertions of quantum physics must be regarded as tentative and that further detailed work will confirm that even the subatomic world is deterministic: it may be concluded that all this, together with the logical implications of quantum indeterminacy, brings quantum physics into the embrace of religious doctrine. Scientists do not all agree and at all times on particular scientific statements or conclusions, which should have been the case if scientific methods and statements were absolutely objective. The disagreement among scientists themselves stems either from the philosophical character of a number of scientific statements, or from the fact that scientific statements raise philosophical questions, which generate responses that may differ among individual scientists. This means that there is some subjectivity in scientific discourse as there is in religious discourse, though the degree of subjectivity is much higher in the latter. This makes the relationship between religion and science one of a closer integration. The fact that there have been innumerable distinguished scientists since the beginning of modern science and even before (such as Copernicus, Galileo, Kepler, Newton, Boyle, John Eccles, John Polkinghorne, Daniel Akyeampong, Francis Allotey) that have had religious faith gives the lie, it seems to me, to the putative conflict or incompatibility between religion and science, or exaggerates the depth and seriousness of the conflict. In these great scientists, religion and science meet; in these great scientists there can be a dialogue – a constructive relationship – between religion and science; in these great scientists there can be an integration of religion and science. It is this foregoing statement that constitutes the foundation for the Kwame Gyekye 19 confident assertion that religion and science will continue to be bed-fellows in the twenty first century and beyond. The reason for this confident assertion is twofold. One reason derives from the wonders of nature. Aristotle asserted that ‘philosophy began in wonder’ (Greek: ek tou thaumazein).30 We can assert, similarly, that religion and science also begin in wonder: in the wonders and mysteries of the created universe that will not cease, in the enigmas and puzzles that constantly beset human life; in the human capacity and the restless spirit of man to wonder, speculate and imagine and, thus, seek to know and apprehend that which is beyond or behind the cloistered walls of man’s limited vision. The other reason for the ever-presence of religion and its unrelenting influence is anchored in man’s awareness of his own limitations – limitations which will ever lead him to search for – and postulate – an unlimited being and seek to do obeisance to this being. Science is a progressive intellectual enterprise and, thus, produces cumulative knowledge and is expected to produce more significant and startling results in the decades to come. I stated at the very beginning of this paper that before science, religion was. I conclude the paper by stating that, in spite of science with its anticipated discoveries in the future, religion will continue to exist, influence and shape the lives of the inhabitants of the world, including scientists, in this century and beyond. 20 Legon Journal of the HUMANITIES Volume 20 (2009) N O T E S S. Sambursky. The Physical World of the Greeks. New York: Colier Books, 1962, 1. p.131. Ian2. G. Barbour. Religion and Science: Historical and Contemporary Issues. San Francisco: Harper Collins Publishers, 1997, p.15. Here and in the next paragraph I have drawn on accounts provided in this book. John 3. Polkinghorne: Science and Theology: An Introduction. London: SPCK/Fortress Press, 1998, p. 6 Ian G. Barbou4. r. Religion and Science, loc. cit. Ian G. Barbour.5. Religion and Science, loc. cit. Ian G. B6. arbour. When Science Meets Religion: Enemies, Strangers, Or Partners? San Francisco: Harper Collins Publishers, 2000, pp.2 -4. Ian G. Ba7. rbour. When Science Meets Religion. p. 28. Kwa8. me Gyekye. African Cultural Values: An Introduction. Philadelphia, PA.:Sankofa Publishing Company, 1997, p. 9. Ian G. 9. Barbour. When Science Meets Religion, p. 70 Ian G. B10. arbour. When Science Meets Religion, P.53 Ian G. Barb11. our. When Science Meets Religion, p. 53. James Trefil. Th12. e Moment of Creation. New York: Collier Books, 1983, p. 223. Daniel A. Akyeam13. pong, Professor of Mathematics at the University of Ghana called my attention to beliefs in deism that arose in the eighteenth century and its implications for religious beliefs. The little I kno14. w of quantum physics is from Ian G. Barbour’s two books referred to above. Ian G. Barbour. Wh15. en Science Meets Religion, pp.67-68. Ian G. Barbour. 16. When Science Meets Religion, p. 68. Daniel A. Akyeamp17. ong. The Two Cultures Revisited: Interactions of Science and Culture. J. B. Danquah Memorial Lectures Series 25, February 1992, Accra: Ghana Academy of Arts and Sciences, 1993, p.20. Daniel A.18. Akyeampong. The Two Cultures Revisited, p. 20. Ian G. Barbour. Wh19. en Science Meets Religion, pp. xiii and 86-87. Kwame Gyekye. An Ess20. ay on African Philosophical Thought. The Akan Conceptual Scheme. New York: Cambridge University Press, 1987; rev. edition. Philadelphia, PA.: Temple University Press, 1995, 9.82. Charles Darwin, 21. quoted in Ian G. Barbour. When Science Meets Religion, p. 10. Kwame Gyekye 21 Personal communication from two of the Physics Professor’s British colleagues in the 22. Physics Department of the University of Ghana some years ago. Karl23. R, Popper and John C. Eccles. The Self and Its Brain. Berlin and London: Verlag Springer International, 1977, p. viii Ian G. Barbour.24. When Science Meets Religion, p. 132. Ian G. Bar25. bour: When Science Meets Religion, p. 130. Ian G. Barbo26. ur. When Science Meets Religion, p.135. Albert E27. instein. “Religion and Science: Irreconcilable?” The Christian Register, June 1948; reprinted in www.google.com: “Religion and Science” p. 9 Albert Einstein. “28. Religion and Science.” New York Times Magazine, 9 November 1930; reprinted in Albert Einstein: The World as I See It.” New York: Philosophical Library, 1949, pp.24—28. See www.google.com: “Religion and Science.” p. 2 Albert Einste29. in: “Science and Religion.” In Einstein’s Ideas and Opinions. New York: Crown Publishers, Inc. 1954, pp.36—40; originally delivered in May 1939. See www.google.com: ‘Religion and Science”, p.5. Aris30. totle. Metaphysics, 982b12 **This paper is a modified version of a Keynote Address to A Conference on Interrelationship between Religion and Science in the 21st Century and Beyond, held at the Kwame Nkrumah University of Science and Technology, Kumasi, 19-22 June, 2008. ** I wish to express my deep appreciation and gratitude to Professor D. A. Akyeampong, Professor of Mathematics at the University of Ghana, whose comments on an earlier draft of this paper I found extremely helpful. I am grateful also to Professor F. K. A. Allotey, a retired Professor of Mathematics and Theoretical Physics at the Kwame Nkrumah University of Science and Technology, Kumasi, who also commented on an earlier version of the paper. 22 Legon Journal of the HUMANITIES Volume 20 (2009) REFERENCES Akyeampong, D.A. 1992. The Two Cultures Revisited: Interactions of Science and Culture. J. B. Danquah Memorial Lectures, Series 25, February 1992. Accra: Ghana Academy of Arts and Sciences. Barbour, I. G. 1997. Religion and Science: Historical and Contemporary Issues. San Francisco: Harper Collins Publishers. Barbour, I. G. 2000. W hen Science Meets Religion: Enemies, Strangers, or Partners? San Francisco: Harper Collins Publishers. Einstein, A. 1930. Religion and Science. New York Times Magazine, November 1930. Reprinted in Albert Einstein: The World as I see It. New York: Philosophical Library. 1949. Einstein, A. 1948. Religion and Science: Irreconcilable? The Christian Register, June 1948. Reprinted in www.google.com Einstein, A. 1954 Science and Religion. In Einstein’s Ideas and Opinion. New York: Crown Publishers. Originally delivered in May 1939. www.google.com. “Religion and Science”, p5. Gyekye, K. 1987, 1995 (Rev. Ed.) An Essay on African Philosophical Thought: The Akan Conceptual Scheme. Philadelphia, PA: Temple University Press. Gyekye, K. 1997. African Cultural Values: An Introduction. Philadelphia, PA: Sankofa Publishing Company. Polkinghore, J. 1998. Science and Theology: An Introduction. London: SPCK/ Fortress Press. Popper, K.R. 1977. The Self and Its Brain. Berlinard London: Verlag Springer International. Trefil, J. 1983. The Moment of Creation. New York: Collier Books. 23 The “Paranormal”: African Philosophy Questions Science Martin Ajei1 Abstract Modern science and its underlying philosophical doctrine, physicalism, have persistently denied reality to a set of phenomena they refer to as “paranormal”. However, belief in the occurrence of these events is common to many non-Western cultures. This essay addresses the question of the reality of these events and advance the view that science, on its own methodology, can neither sustain the denial of their reality nor justify its rejection of the “paranormal” as sources of knowledge. 1. Introduction This paper attempts a critical review of the dismemberment by physical science and its underlying philosophical doctrines of the set of phenomena referred to as “paranormal”. A widely accepted definition of a “paranormal” event is that which seems to contradict the fundamental ideas and principles upon which modern science has been based (Mosley1978: 9). Belief in the occurrence of these events is common to many non-Western cultures, yet their ontological status and, with it, the question of whether they can be legitimate sources of knowledge has occasioned academic debate since the end of the nineteenth century1. All subscribers to physicalism, the ontological and epistemological framework which sustains mainstream Western science, have persisted in their denial of existence to these phenomena and the claims to knowledge made by their advocates. This essay will address the question of the reality of ‘paranormal’ events and defend the thesis that they are real and can furnish us with legitimate knowledge. Therefore science, on its own methodology, cannot sustain the denial of reality to the ‘paranormal’. Nor is the rejection of the ‘paranormal’ as sources of knowledge methodologically tenable. Toward this, we expound indigenous African ontological and epistemological concepts to explain the ‘paranormal’. The physical scientist who denies paranormal phenomena must provide better explanations than those offered by indigenous African knowledge systems. We claim, further, that the principles underlying the methods of scientific practice, interpreted more broadly than that commonly understood in the West, are embedded in the philosophies of many traditional African cultures, and that certain African indigenous concepts become viable scientific postulates for the explanation of the ‘‘paranormal’’. “Sunsum” an Akan ontologico-epistemological concept is a viable “scientific” postulate for the 1 Dr. Martin Ajei is a Lecturer in the Department of Philosophy, University of Ghana, Legon. 24 Legon Journal of the HUMANITIES Volume 20 (2009) explanation of the ‘paranormal’. Therefore, broadening the methodology of science to embrace the procedures that traditional African thinkers employ for the investigation of nature can make meaningful contributions to the search for knowledge and truth. The argument for the broadening of the methodology of conventional science speaks to the need to take African, Latin American, Asian and Australasian indigenous knowledge systems seriously. Our discussion falls accordingly within the general context of indigenous knowledge systems and, in particular African indigenous knowledge systems with particular reference to Akan cosmology. It is not our intention to engage in an elaborate disquisition on the meaning of the phrase, African indigenous knowledge systems. Although such a special discussion is scientifically apposite, we shall content ourselves with the acceptation that the phrase refers to the knowledge, beliefs and cultures of peoples whose ways of knowing and acting were and, continue to be systematically discarded – in the name of “science” and “civilization” – from the global construction of “scientific” knowledge. 2. Occurrence of the “Paranormal”: Myth and Reality The founding of the Society for Psychical Research (SPR) in London in 1882 is often taken to mark the first effort in the Western world to address the question of the occurrence of paranormal phenomena scientifically. The Society undertook to “investigate, among other things, the alleged acquisition of knowledge without the use of any known sense organ” (Mundle 1967: 49-50), and the commonest term given to the kind of information identified for investigation on the agenda of the SPR is Extra-Sensory Perception (ESP) or Paranormal Cognition (PNC). The terminologies employed by the Society in defining the subject of investigation are instructive on the skepticism about the occurrence of such phenomena. The empiricist stance permeates the terminologies. The aim of the society suggests clearly that the sense organs (and elliptically the pure activity of the mind processing the impressions of the sense organs) constitute the pathways to the acquisition of a class of knowledge about which there is no allegation of unreality. These then, by implication, must be the normal pathways to knowledge. There is, with this, the implication of the existence of a possible category of knowledge, that which is alleged to exist para the normal, and whose status must be judged according to the standards that determine the normal. Embedded in this conception is that the events which are para the normal can never fully coincide with normal events. It is not unreasonable to suppose that this prejudgment in defining the research tasks of the society have provided a conceptual background that excluded Martin Ajei 25 and clouded alternative viewpoints and nourished the self-assurance of critics of the paranormal. This is the reason why we use the term “paranormal” in quotation marks. Experimental study of the “paranormal” has since then been the approach of parapsychology or psychical research, and by the 1940s, experimental study had accumulated enormous evidence for the occurrence of such phenomena in Europe and America (Carrington 1945; Johnson 1955; Rhine 1934; Schmidt 1969; Stanford 1978; Tyrell 1953). The researchers identified two broad categories of “paranormal” events. These are Paranormal Cognition (PNC) or Extra-Sensory Perception (ESP), on the one hand, and Psychokinesis on the other. They distinguished four main sub-species of PNC, the first of which is telepathy, a term they employed to refer to the communication between one mind and another without the use of the recognized channels of sense. In ordinary parlance, we identify telepathic ability as the ability to ‘read someone’s mind’. Next is clairvoyance. This refers to the mind’s ability to acquire knowledge of physical objects or events extra-sensorially. An example of clairvoyant ability would be X’s ability to state accurately the contents of a closed briefcase that he encounters for the first time. Precognition, the third species of PNC, refers to the ability to acquire fore-knowledge of events without the use of recognized sensory channels; and finally retrocognition refers to the non-inferential cognition of events in the past, verified later, which are outside the range of a person’s memory (Johnson 1955: 9 - 57). On the other hand, all the phenomena in which the mind is deemed to have extra-sensorial powers of action in the world constitute psychokinesis. The most important sub-species of this, for our purposes, is witchcraft, since this constitutes a good portion of the belief in the “paranormal” in Africa. Witchcraft can be defined as an innate power which can be used by its possessor to influence events in the world from any distance (Oluwole 1995: 366). 3. Skepticism of the “Paranormal” In spite of the available evidence of experimental research, subscribers to physicalism, the philosophical doctrine underlying the physical sciences, have deployed diverse claims to discredit the findings of psychical researchers and condescend on the conclusions of non-western cultures that maintain the occurrence of events that do not easily lend themselves to the explanatory framework of physicalism. Most of this skepticism can be brought under four main categories. (a) Accidental occurrence Perhaps the most common of these is the chance hypothesis. This is the claim that “paranormal” events have occurred by persistent accident. This, of course, is a 26 Legon Journal of the HUMANITIES Volume 20 (2009) self-refuting argument for the champion of this hypothesis because an accidental event is an event, and one counter-instance is sufficient to defeat pretences of universalism. Besides, the chance hypothesis amounts to the claim that evidence stretching through varied conditions in different cultures over a wide range of subjects at different times is fortuitous and therefore unreasonable. These notwithstanding, sufficient documentation exist to prove the conformity of the methods of psychical researchers with established scientific methods of experiment and analysis of testimony. To ensure the validity of their findings, the psychical researchers employed the mathematical evaluative principle of probability, and their calculations of chance, from the 1920s onwards, fulfill the requirements for general practice among statisticians (Rhine 1934: 150). In 1937, the American Institute of Mathematical Research affirmed that these statistical analyses were essentially valid (Mundle 1967: 55), and recent meta-analysis of these findings supports this (Utts 1991). What these mean is that scientific method has established the occurrence of these events. Consequently, pleading the chance hypothesis, by employing scientific methods as the measure of the reality (i.e. the non-chance occurrence) of a phenomenon, leads to inconsistency and cannot therefore stand. Once the chance hypothesis is ruled out further questions arise regarding human reliability. These are whether we can vouch for the sincerity and competence of the researchers, and if yes whether loopholes may have existed in their techniques that might annul their conclusions? (b) Sincerity The subject of sincerity requires very little in answer. Psychical research, like most academic pursuits, is primarily a quest for truth about nature which, for many decades, was pursued by “several university departments in most of the [western] world” (Koestler 1974: 15). Therefore, a suspicion of fraud mars the integrity and probity of thousands of respectable scientists. But it would be unreasonable to suppose the reality of a conspiracy on such a scale. Hence the fraud hypothesis must be dropped. The remark of J. B Rhine, a psychologist who led a para-psychological laboratory in Duke University for forty years, on the question of incompetence is that: “It is also probable that some errors have been made in recording, totaling and computing values. If so, such errors are at most of trifling consequence. The general ground has been covered too often and by too many individuals for serious error that would vitiate an important conclusion” (Rhine 1934: 154). In other words disbelief in the competence of the researchers can be sustained only by a fanatic and unintelligent doubt. Therefore doubts based on both the fraud and incompetence hypotheses are deemed to be unjustified. Accordingly, they must be Martin Ajei 27 rejected. Even here it is worth noting that the fact that a scientist vouches for his or her competence does not by necessity exclude the possibility of error. The claim to competence is not the affirmation of infallibility. (c) Scientific proof The remaining source of skepticism of the reality of “paranormal” events worth considering is that these events are difficult to confirm in a way that meets the demands of proof of science. As with the others, this claim can hardly be rationally defended. The statistical nature of scientific justification as the basis for strict replication and prediction is a necessary but not sufficient condition for recognizing the objectivity of science (Doyal and Harris 1986: 52-72). In upholding ‘objectivity’ science necessarily excludes certain elements and segments of reality from its purview. It is therefore narrow and limited in scope. Elements or segments of reality falling outside the scope of science are no less real because of their exclusion. Now, a review of PNC experiments reported since 1934 indicates a statistically significant replication rate of sixty one percent of approximately 3.3 million trials (Honorton 1978: 487). Honorton observes that “this is 60 times the number of significant studies we would expect if the significant results were due to chance error (Honorton 1978: ibid)”. It is reasonable, then, to view the replication status in the findings of psychical research within the larger perspective of replication demands in other areas of scientific research. It becomes clear, when this approach is adopted, that ‘paranormal’ phenomena are not the only events in which strict replication is unfeasible. In the medical and behavioral sciences, the demand for strict replication is an exception to the rule (Beloff 1994: 9). Even in contemporary theoretical physics, the ability to predict future events under definite circumstances seems elusive. Thus “Yes! Physics has given up. We do not know how to predict what will happen in a given circumstance, and we believe now that it is impossible – that the only thing that can be predicted is the probability of different events” (Feynman, Leighton, & Sands 1965: 10). Therefore, the challenge of non-replication may also be set aside. The conclusion to be drawn from this is that the recalcitrance of the subscribers to physicalism must be solely because these events defy their explanatory framework. It remains to be seen how justified this recalcitrance is. Before we attend to the question of this justification, we turn to some features of the ontological and epistemological systems of African cultures, which do not fully endorse “the fundamental ideas and principles upon which modern science has been built”, and consider how they legitimately sustain the reality and rationality of the “paranormal”. 28 Legon Journal of the HUMANITIES Volume 20 (2009) 4. Foundations of the belief in the “Paranormal” in African Philosophy The fundamental structure of Akan cosmology and ontology is fairly well documented (Ajei 2000; Busia 1963; Danquah 1968; Gyekye 1995; Rattray 1927) and the predominant interpretation of these by traditional Akan sages, philosophers and anthropologists alike reveals the following three essential features: The first is the postulate that existence comprises visible and invisible realms. Secondly, Akan thought postulates a universe containing a hierarchy of beings, with Onyame (the Supreme Being) at the top. In descending order from Onyame comes the abosom (deities) the Nsamanfo (ancestral spirits) human beings and physical objects. Some commentators have attributed a dualistic universe to the Akans (Gyekye 1995: 87), assigning all the categories of being beyond human beings to be the immaterial/spiritual realm of existence, whereas observable entities like trees have been consigned to the physical realm. We would like to suggest here that although the Akan thinker affirms that the universe is composed of visible and invisible beings, it is doubtful whether the implication is that these aspects of existence are two distinctly separate categories as the notion of dualism would suggest. Rather the Akan thinker conceives of these two realms as two points on a continuum, and not distinct realms. This makes a clear-cut dichotomy of the immaterial/material or spiritual/physical in Akan ontology implausible. The third essential postulate of Akan cosmology is that the universe is endowed with varying degrees of force of power, all of which derive ultimately from Onyame. This force or power is sunsum, “a universal spirit, manifesting itself differently in the various beings and objects in the natural world” (Gyekye 1995: 73) or “a general power to act in non-ordinary, non-physical ways” (Minkus 1977: 115). In particular existents, sunsum itself manifests as an activating essence2. On this view, all existing things are similarly unified by their indispensable possession of sunsum, a universal force that functions as the activating principle of their being. The concept, then, accounts for the basic similarity of all existents, which are alike in being spirited or active. But what is meant by the claim that sunsum is a ‘power’ or ‘force’ that pervades the universe? Most traditional Akan sages and modern philosophers alike admit that sunsum, as force or power, is locatable in this world and is accessible to sense experience under certain conditions. It is immediately concealed from the senses, but the shroud that veils sunsum from immediate perception may be removed under certain conditions, which may be achieved by training or by the conscious (or unconscious) assumption of a certain mental state (Ofosu: 1999). As a force capable of manifesting, and influencing events, in the physical world, sunsum, by nature, is susceptible to the system of coordinates that define the spatio-temporal schema. Yet, it is not like the concepts of mainstream Western Martin Ajei 29 philosophy and science, guided exclusively by the canons of the logic of binary opposition with its central principles of Identity and Non-Contradiction. This is the case because this logic, primarily concerned as it is with the form rather than the content, of thought, is ill equipped to capture the content and all the ramifications of the dynamism of a concept like sunsum. In order for an existing thing to be active or have consciousness its sunsum has to interact with the energy of something else in the universe. Consequently, philosophers in the indigenous African setting define experience as the outcome of interaction of sunsum. These interactions lead to events of which we may be cognitively, affectively or sensorially aware, and sometimes not. The vitality of the sunsum possessed by the various beings on the hierarchical structure of Akan ontology differs, and these different levels of potency of sunsum determine the levels of awareness of these beings. If the ability to consciously experience an event depends on the vitality of one’s sunsum, then a human being may be endowed with just sufficient sunsum to enable him/her consciousness only of the physical world, or some aspects of it. However, this vitality may be augmented to enable a person attain ‘cosmological consciousness’, a level of conscious energy that is capable of apprehending “our understanding of the divine and our relationships to the divine and everything’s relationship to everything else” (Grills 2002: 12). Cosmological consciousness, then, denotes the ability of a person’s sunsum to maximize the use of primordial intelligence with which he was endowed at birth, permitting it consciousness of events which are not ordinarily knowable to others. The Akan thinker’s notion of sunsum as force coincides with the notion of energy in Western science, conceives of energy as all that was, is and ever will be (i.e. something that can be neither created nor destroyed). On this view, energy may not be visible, but we can understand its properties to do useful tasks called work. The transferring of energy from one object to the other is what is meant by work. Therefore, work is the product of a force (F) times a distance (d). Newton’s First Law states that if there is no net force on a body, then that body must remain at rest if it is initially at rest. It is important to note two things at this juncture. First, that Western science has for centuries admitted as one of its most important theoretical postulates something (energy) that cannot be seen by our ordinary channels of perception, but experienced. Secondly, that an energy source can neither be created nor destroyed. 30 Legon Journal of the HUMANITIES Volume 20 (2009) 5. Wholeness and Sunsum Due to the belief in the existence of a universal active force that derives from Onyame, the Akan thinker can legitimately be held to conceive of being or nature as one. In this ontology, human beings, like all categories of being partake, by virtue of the species of sunsum in them, of the being of Onyame. Therefore, we cannot plausibly separate being (as matter) from being (as consciousness). Being is understood as being wholeness. I partake of sunsum and so does that tree. Our sunsum derives from one source. Since all the beings in a wholistic framework share characteristics of the various aspects of a complex whole that forms an internally contiguous order, holonistic ontology succeeds in harmoniously fitting the multiplicity of beings in a single domain of existence. Now, it is important to understand that in this oneness the essential relations among the diverse beings allow for such fluidity that the categories which define their essential attributes dovetail into each other (Abraham 1962: 51). What this notion of beings perpetually emanating from, and gravitating towards the source of their existence means is that their existence is merely a pattern of interactions within a wholeness that each existing entity has the power to interact with every other entity and, as such, is a key to universal knowledge. A holonistic conception of the universe therefore affords us a picture of nature as a multiplicity of entities that interact in both determinate and indeterminate patterns. It is a view that furnishes us with a language for describing or reasoning about natural and man-made phenomena in a way that both acknowledges the importance of logic and the limitations imposed upon its predilection for form rather than content. It is necessary to transcend these limitations on the ground that actually, matter and form are indivisible. This indivisibility speaks to the need to recognize the emergence of a logic consistent with it. The language of the indivisibility, in actu, of content and form reveals the ability of human conscious energy to operate simultaneously at various levels of existence, of the human capacity to interact with other entities in nature whose being humans cannot accurately define. The logic consistent with this conception may be referred to as the rheomode language (Ramose 1999: 56-60). In this order of things, our search for unity (for a definite, absolute knowledge of nature) reveals that no method of investigation can yield a permanent truth about nature. Instead, we are encouraged to recognize and to explore the interdependence of significant elements in a totality, and to appreciate the relevance of this to our knowledge about nature. How, then, are events caused in this holonistic cosmology and ontology of African thought? Martin Ajei 31 6. Causality in a Holonistic Framework and the Justification of the “Paranormal” Many African philosophers are agreed that indigenous African cultures subscribe to the principle of universal determinism. Thus “when a European explains an unpredictable or unexpected natural event by reference to chance, coincidence, luck, or fortune, from the point of view of Akan thinkers that is the same as saying that the cause is unknown. But the Akan thinker would here retort that ignorance of the cause of an event does not imply the non existence of a cause” (Gyekye 1995: 82). The determinism upheld by African thinkers differs in a fundamental manner from that of Western philosophy and physical science. Embedded in the Western notion of causation is the idea that a cause (Y) must necessarily precede the event (X) which Y causes. There may be an intermediate chain of events between X and Y, but all these must progress from Y before they can properly be considered as part of the cause of X. Within the deterministic universe of Africans, however, there are two types of causal explanations. The first type pertains to what Gyekye describes as “Why 1” questions. In these, contiguous causal relations are established and therefore physical laws are considered relevant and sufficient to explain relations between phenomena. This is the same as the theory of causality upheld in Western philosophy and science. But where the events to be explained are considered extraordinary in character, such that physical laws prove inadequate to explain them, the investigator would resort to a “Why 2” question and its attendant explanatory model. By so doing, the investigator admits the limitations of the explanatory capacity of physical laws. And in a “Why 2” explanatory model, the interplay of sunsum becomes the fundamental postulate. This two-pronged causal theory is common across African cultures. Sogolo, for example, refers to this as “primary” and “secondary elaborations” with regard to causality in African thought (Sogolo 1993: 91-116). Writing of the Yoruba concept of Inner Essence, Ayoade suggests that we should not consider non-supernatural and supernatural questions as constituting two irreconcilable categories of causal theory but rather as different points along the same continuum. He writes “this is particularly so because the Yoruba believes that a non- supernatural ailment makes a patient highly susceptible to, or softens up a patient for, the infliction of a supernatural ailment” (Ayoade 1979: 49). Ayoade’s ‘non- supernatural’ and ‘supernatural’ is coterminus with Gyekye’s ‘Why 1’ and ‘Why 2’. Therefore in Yoruba causal theory, Gyekye’s Why 1 and Why 2 types of causes would be complementary and mutually inclusive. It is tempting to pursue this line of thought with particular reference to the classical discussion on the principle of 32 Legon Journal of the HUMANITIES Volume 20 (2009) sufficient reason. Leibniz is a well-known proponent of this principle (Parkinson 1965: 56-75; Russel 1937: 30-35). Without succumbing to this temptation, we suggest that this principle is, on the basis of a critical discussion, amenable to the African conception of causality. From his analysis of an invented fire brought about by an arsonist, Sogolo affirms this view that different causal explanations may affirm and complement each other by advancing the view that a given event can have a variety of causal explanations and our interest in that event determines which of these various causes we advance (Sogolo 1993: 107-108). One of the suggestions to be inferred from this is the importance of cultural determinants in the analysis of ‘cause’, a position curiously compatible with Hume’s (1975) attack on the concept of causation.3. Since this attack, it is now largely accepted by both philosophy and science that a constant conjunction between two events is sufficient to prove a ‘causal connection’ between them. This casts doubts on the legitimacy of the demands for an objective, value-free and universal notion of causality, especially in the behavioral sciences. If the explanation of an event is legitimately determined by my interests, then I am surely providing sufficient explanation for the event by claiming that X caused it so long as I observe a constant conjunction between the event and X. This integrated approach to causal analysis makes the explanation of the “paranormal” by African ontologies, plausible. This is because in explaining the events which pose anomalies and paradoxes to Western science, this approach proceeds from the knowledge that although physical laws make for regularity in the explanation of nature, they cannot exhaust all possible explanations of events. This is because for the African the ontology of invisible beings is as real as the reality of the observable physical world. The natural laws of physicalism, which presuppose regularity in nature, become inadequate because such regularity is undermined by the existence of irregular events that also form part of human experience. For the Akan and, by extension the African, the cause of these latter events is an invisible force present in the world, sunsum. It is difficult to understand why physical scientists should have a problem with this, because in fact Newton’s lifelong dissatisfaction with gravity4, was because of his understanding that gravity was paradigmatically action at a distance, a mode of causation which seemed to have no place in his comprehensive system of mechanics5. In other words, Newtonian physics, which has shaped the theories and practice of science from the seventeenth century onwards admitted at least two characteristic modes of causality. Martin Ajei 33 7. The activity of sunsum in a holonistic framework The organism, then, constitutes the minimum entity of analysis in a holonistic framework. For, although the event, say X, may consist of parts, all these parts become modes in which they (the parts) enter into the composition of X. What, then, would be the place of sunsum in this schema? Its role would be to link the singular events not a unity, and this role becomes particularly evident when, for instance, a diviner performs an act of precognition which brings together parts of reality that apparently have no link. But how does sunsum serve this function? If we grant the hypothesis that sunsum is a force that inheres in every existing individual as well as pervading the space outside individuals in the perceptible world, then it is not difficult to accept that the sunsum dwelling in a particular existent is capable of escaping its spatial localization and regain its membership of the universal sunsum. The localized sunsum is only a species of a universal phenomenon. In a holonistic conception of nature, the activity of sunsum can provide us with a meaningful explanation of precognition because the structure of space and time in this conception is different from the physical-scientific conception of it. The potency of the diviner’s psychic power (his personal sunsum) enables him to collapse the barriers erected in our consciousness by our fixed conception of space and time that normally serve as indispensable coordinates for our understanding of events. By this collapse of space and time, the diviner’s personal sunsum becomes a receiver in the present of events ‘transmitted’ from the future. Sunsum as energy is the medium of this transmission. The potent psychic energy of the diviner (personal sunsum) apprehends the future event through the medium of universal sunsum which pervades space and time. It is because of the wholeness that universal sunsum can serve as the vehicle through which ‘the whole situation’ can be grasped by a potent personal sunsum. For the diviner, therefore, something unknowable in the physical scientific framework – the future – becomes actually known because the barrier (time) that shields the future away from the present is removed. Through the sunsum the “supernatural” is one with the natural world. In this connection it is necessary to inquire why telepathy and fortune telling seem to be so popular in the West despite its skepticism about the “paranormal”. Furthermore, the oneness of the universe, according to Akan cosmology, raises the interesting question of whether or not African philosophy has only an ontology and, not an ontology and a metaphysics. It is outside the scope of this paper to consider this question. The sunsum that pervades space outside of individuals provide a link by which individuals can be connected, although they may themselves inhabit a different location in space. This link may be contiguous, but because sunsum is inaccessible to ordinary sense, our normal perceptive channels are unable to 34 Legon Journal of the HUMANITIES Volume 20 (2009) apprehend this contiguity. It is in this connection that sunsum explains adequately the “paranormal” events and substantiates its presence in our human sensible dimensions of space and time. 8. Empirical justification for the occurrence of the paranormal outside of the experimental setting The hypothesis of sunsum as an active cause of phenomena has an empirical basis in the sense that its proponents regard the occurrence of these phenomena to be adequately established either by observation or testimonial evidence (Engmann 1992: 186). Our interaction with thinkers in the traditional setting also confirms their willingness to relate theory to facts, in the sense that they infer the existence of the hypothesized force from concrete manifestations that they attribute to it, and consider these manifestations sufficient reason for their inference. Oluwole argues for the validity of this inference by contending that the positive result of any one of the following would be sufficient to establish the reality of witchcraft: we may be given an explanation of the i) modus operandi of witchcraft power we may experimentally establish a causal relationship between t h e ii) postulated paranormal power and the event that proves the practical efficacy of this power we may show a practical manipulation of the power (Oluwole 1995: iii) 367) Oluwole then submits that the African’s claim to the rationality of her belief in the reality of “paranormal” forces rests in her ability to infer their activity by virtue of (ii) and (iii) above. Thus the African establishes certain facts as the effects of something (a force) whose existence may not be denied precisely because of those facts and, on the basis of African ontology, infers sunsum as the cause of those facts. This inference is strengthened by his/her knowledge of the abundant testimony to the apparent ability of diviners and sangomas to control and manipulate these powers. On this view, then, the African’s belief in the “paranormal” explains witchcraft. The explanation boils down to a statement about the demonstrable effects of this belief in practice. The rationality of the belief in sunsum as a theoretical positum on the basis of an ontology is well within the purview of scientific practice. In fact purely objective knowledge, if by that is meant knowledge in which metaphysics – though from an African philosophical standpoint ontology of invisible beings is preferred to metaphysics – does not play a part, cannot exist. A “metaphysics” and values are necessarily prior and foundational to every science, conceived as a body of Martin Ajei 35 knowledge and practice6. Two special and striking cases of the profound influence of “metaphysics” on the pursuit of science in the Western experience deserve brief mention here. The first is that Locke’s metaphysical doctrine of primary and secondary qualities7 provided the basis for Newtonian physics (Mackie 1976: 22-23). The second is Einstein’s Special Theory of Relativity, the metaphysical foundation of which is the claim that any system of objects has equal title to the claim to be ‘absolutely at rest’ and hence any of them can be chosen as the frame of reference by which the motions of all other systems are determined (Harre 1972: 15-16). Consequently, the legitimate ideals of objectivity are better achieved by understanding that when particular theories and assumptions prove to be a bar to progress in understanding the nature of the world, they are liable to be changed or improved. Thus “the assumption underlying the current requirements of empirical verification [by Western Science], the assumption that information can be transmitted and received by human beings only through the recognized channels of the senses, is subject to amendment in the future” (Oluwole 1995: 368). It is not insignificant that Burtt (1932) has traversed this territory through his historical and critical essay under the title, The Metaphysical Foundations of Modern Physical Science. 9. The Endorsement of “Paranormal” Cognition by the Causal Theory of Knowledge Plato’s account of knowledge in his Thaetetus was widely accepted as sufficient in the Western philosophical setting until it was challenged by Gettier, who provided examples to show that the justification principle in Platonic epistemology could produce a situation in which a false belief is (or can be) justified (Gettier 1963: 122-123). The problems posed by Gettier inspired a set of approaches to the analysis of knowledge, one of which is the Causal Theory which, we believe, affords a means by which “paranormal” cognition could be validated. The idea underlying all the various formulations of the Causal Theory is that the justification condition for knowledge is met if one’s belief is caused in an appropriate way (Appiah 1989: 52). But what precisely is appropriate, and how do we decide on that? A widely acclaimed interpretation of “the appropriate way” of getting a true belief is to be able to distinguish the actual state of affairs in which the belief is true from relevant possible states of affairs in which the belief is false. This discrimination of the actually true from the possibly false state of affairs is best done by a method that is reliable in the circumstances (Goldman 1978: 124). 36 Legon Journal of the HUMANITIES Volume 20 (2009) A few perspectives emerge from Goldman’s rendition of the justification condition that supports our claim that Causal Theories validate “paranormal” cognition. The first is the suggestion that a reliable method for distinguishing true beliefs varies with the circumstances. This means simply that what constitutes good evidence for a true belief is relative to the state of affairs prevalent in the belief situation. Since a reliable method justifies a belief, according to the Causal Theory, and since much in the foregone supports the claim that belief in “paranormal” cognition can be shown to be based on ‘a reliable method’, then belief in “paranormal” cognition is rationally validated by the Causal Theory. 10. The African Diviner as a challenge to “science” The validation of “paranormal” events by a theory in mainstream Western epistemology and science leads us to develop the argument, advanced earlier on in this paper, that on the basis of developments in some branches of mainstream Western science, the African holonistic practitioner of the “paranormal”, who identifies, accesses and harnesses the potency of sunsum for action in the world, can share the status of ‘scientist’ in a redefined concept of science. The dominating note of science until the beginning of the 20th century appears to be that the metaphysical doctrine of physicalism affords an adequate basis for scientific concepts. Many complexities have developed in respect of the fundamental concepts of science such as matter, space, time and energy. These have undermined the assumptions of physicalism. The result is that it can no longer hold. Consequently, its relevance to scientific knowledge has been profoundly diminished. The factors that first led physicists to distrust their orthodox faith were two theoretical systems developed between 1900 and 1927. One was quantum theory, dealing with fundamental units of matter and energy, and the other was the Theory of Relativity, that deals with space, time and the structure of the universe as a whole. Both of these theories are now accepted pillars of contemporary scientific thought. We propose to concentrate on quantum theory here because its findings provide better analogies between the foundations of the theories and methods of the practitioners of the “paranormal” and contemporary physicists. At the heart of quantum theory is the denial that an atom really exists as an independent entity (Davies 1990: 72-90). If it does, then at the very least it should have a location and a definite motion in space at all times, and this should apply to its components as well. But the findings of Quantum Theorists suggest that it is impossible to know simultaneously the velocity and the position of an electron orbiting around an atomic nucleus (Sklar 1992: 164-179). This is the celebrated Martin Ajei 37 Principle of Uncertainty, formulated by Heisenberg, one of the founders of this theory (Heisenberg 1962: 44-45). It states simply that it is impossible with any of the principles now known to science to determine the position and the velocity of an electron at the same time – to state confidently that an electron is “right here at this spot” and is moving at such and such a speed”. This is because micro particles have been observed to traverse their path in space discontinuously: they move to other parts in space only by disappearing and then reappearing at a new location: they change location by “quantum jumps” (Powers 1982: 138-152). The implications of this principle for us are these: (i) if there exists no exact state of the particle, the idea that a real thing is that which can be located at a certain place at a certain time is questionable and, with it, the classical notion of causality. (ii) If all explanations and predictions of micro phenomena can only be indeterminate, and it is these micro phenomena that are the ultimate constituents of macro phenomena, then the cause of events and phenomena at the macro level must be indeterminate as well. Where then do we stand with regard to the classical notion of causality? To give meaning to the wave-like and particle-like representations of nature, Niels Bohr proposed another principle: the Principle of Complementarity, which states that the particle picture and the wave picture are two complementary descriptions of the same reality, each of them only partially correct and only within a limited range of application. For a full account of reality both viewpoints are needed, even though they are incompatible and cannot be viewed simultaneously (Akyeampong 1993: 20-21). What this principle seems to suggest, then, is that the metaphysical barrier that separates the human observer from ‘objective reality’ has become obscure. Further, it supports the view that there exist phenomena in nature which cannot be explained or understood with the current methods of science: since this principle rejects the notion that a particular explanatory framework can be absolutely dependable, it establishes, by implication, the necessary limitation of the insights of physicalism. It seems to advocate a complementary and inclusive approach to the study of reality, affirming the perspective that the cumulative yield of several points of view will produce a better understanding of reality than any single framework of knowledge. Both the Complementarity and Uncertainty Principles substantiate the plausibility of our elucidation of the concept of sunsum. We maintain that a particular sunsum is a species of the universal sunsum, and that for this reason it is capable of transcending our normal conceptions of space and time. Furthermore, the Uncertainty Principle invalidates the classical notion of causality since the 38 Legon Journal of the HUMANITIES Volume 20 (2009) location of the causes of events in the micro world cannot be completely determined. Similarly, the African cosmologist postulates a causal element (sunsum) whose location and mechanisms are indeterminate. Thus at the dawn of the 21st century, a new scientific worldview has emerged in which the objectivity of the natural world has been replaced by the view that the behavior of the object under investigation changes as we change our point of view. The present scientific worldview, especially the Complementarity Principle, seems to agree with the notion that there are several mutually exclusive approaches to reality. It confirms the view that there are phenomena or experiences in nature that cannot be understood or explained with the methodology of physicalism alone. 11. Conclusion We have shown that the principle of causality in Western science remains responsible for the exclusion of the “paranormal” from the purview of science. Yet, the principle of causality is itself seriously vitiated by Heisenberg’s indeterminacy principle on the one hand and Bohr’s complementarity principle on the other. Accordingly, the weak foundation upon which the principle of causality rests cannot be definitive justification or denial of the “paranormal”. On this basis, the Akan concept of sunsum having strong conceptual and cultural affinity with moya, moea, umoya (Ramose 1999: 85) in the Bantu languages, has been invoked in support of the thesis that the unduly narrow and restrictive concept of science upheld by the West must be broadened in order to sustain its claim to science. Such broadening must also mean the recognition of sunsum and other closely related African concepts as scientifically viable. Without such recognition both the credibility and the tenability of Western science shall remain in doubt from the point of view of African indigenous knowledge systems. Martin Ajei 39 NOTES This is exemplified by the establishment of the Society for Psychical Research in 1. London in 1882 This is affirmed2. by both Gyekye, (1995: 72) and Minkus, (1977: 114). What this means is that the universe of traditional Akan thought contains no inert objects in the sense of matter incapable of awareness or action. This notion that every existent is composed of an activating principle constitutes a difference between the metaphysic of the Akans and other metaphysical systems. It rejects by implication Cartesianism and other all metaphysical systems that hold that matter is essentially passive or inert and that a creative divine being must therefore activate it In his Enqui3. ries, Hume argues that the physical existence of causal necessity is demonstrable neither on purely rational nor empirical grounds. This, according to Hume, is the case since no logical contradiction is involved in assuming that the relevant correlation between ‘cause’ and ‘effect’ will cease to occur: “when we look about us towards external objects, and consider the operation of causes, we are never able, in a single instance, to discover any power or necessary connection; any quality which binds the effect to the cause, and renders the one an infallible consequence of the other. We only find that the one does actually, in fact, follow the other. The impulse of one billiard ball is attended with motion in the second. This is the whole that appears to the outward senses. The mind feels no sentiment or inward impression form this succession of objects: consequently, there is not in any single, particular instance of cause and effect, anything which can suggest the idea of power or necessary connection.” (p. 63) In his Papers and Le4. tters on Natural Philosophy, Newton was at pains to explain that he did not consider gravity to be a primary power of matter since it acted at a distance. Newton’s laws of impac5. t describe the redistribution of velocities and changes in direction that takes place when two or more things collide. This is the very essence of Action by Contact. This claim alludes6. to Kuhn’s (1970) subsuming of all scientific activity under a paradigm. And a paradigm, according to him, consists of all the theoretical and methodological assumptions and operations which operate under a period of ‘normal science’ during which scientists do not disagree about their conclusions once they 40 Legon Journal of the HUMANITIES Volume 20 (2009) are derived and tested from within the prevailing paradigm. The claim can also be attributed to Mudimbe, 1988; Harding, 1997; and Ramose, 2002. According to whic7. h certain qualities of bodies were held to be known to us in sensory experience while others manifest in the things themselves. Martin Ajei 41 REFERENCES Abraham, W. E. 1962. The Mind of Africa. Chicago: University of Chicago Press. Ajei, M. 2000. The Paranormal: An Inquiry into Some Features of Akan Metaphysics and Epistemology. Unpublished M.Phil Thesis Submitted to the Department of Philosophy, University of Ghana. Akyeampong, D. 1993. The Two Cultures Revisited: Interactions of Science and Culture. Accra: Ghana Academy of Arts and Sciences. Appiah, K. 1989. Necessary Questions. Englewood Cliffs: Prentice-Hall. Ayoade J. A. A., 1979. The Concept of Inner Essence in Yoruba Traditional Medicine. In Ademuwagun, Z. A. et al., (eds), African Therapeutic Systems, Waltham, Mass: Crossroads Press, pp. 49-55. Beloff, J. 1994. The Mind Body Problem. The Journal of Scientific Exploration: 8.4.Burtt, E.A. 1932. The Metaphysical Foundations of Modern Physical Science. London: Routledge & Kegan Paul. Busia, K. A. 1963. The African World-View. In J. Drachler (ed). African Heritage. New York: Crowell, Collier & Macmillan. Carrington, W. 1945. Telepathy. London: Methuen and Co. Danquah, J. B. 1968. The Akan Doctrine of God: A Fragment of Gold Coast Ethics and Religion. London: Frank Cass. Davies, P. 1990. God and the New Physics. London: Penguin Books. Doyal, L. & Harris, R. 1986. Empiricism, Explanation and Rationality – An Introduction to the Philosophy of the Social Sciences. London: Watts & Co. Engmann, J. 1992. Immortality and the nature of Man in Ga Thought. In Gyekye, K. & Wiredu, K. (eds.). Person and Community: Ghanaian Philosophical Studies I. Washington D, C.: The Council for Research in Values and Philosophy. Feynman, R. P., Leighton, R. B., & Sands, M. 1965. Quantum Behaviour, In The Feynman Lectures on Physics, Vol. 3, Quantum Mechanics. Menlo Park: Addison-Wesley. Gettier, E. L. 1963. “Is Justified True Belief Knowledge?”. Analysis: 23. 121-123. Goldman, A.1978. Discrimination and Perceptual Knowledge. In Essays on Knowledge and Justification. Pappas, G. S. and Swain, M (eds.). Ithaca: Cornell University Press. Grills, C. 2002. African-Centered Psychology : Basic Principles, in T. Parham (ed). Counseling Persons of African Descent. Thousand Oaks: Sage Publications. 42 Legon Journal of the HUMANITIES Volume 20 (2009) Gyekye, K. 1995. An Essay on African Philosophical Thought: The Akan Conceptual Scheme. Philadephia: Temple University Press. Harre, R. 1972. The Philosophies of Science. Oxford: Oxford University Press. Heisenberg, W. 1962. Physics and Philosophy. New York: Harper & Row. Honorton, C. 1978. Has Science Developed the Competence to Confront Claims of the Paranormal?, In M. Ebon (ed.), The Signet Handbook of Parapsychology, New York, New American Library, p 487. Hume, D.1975. Enquiries Concerning Human Understanding. Oxford: Clarendon Press. Johnson, R. C. 1955. Psychical Research. London: English Universities Press. Koestler, A. 1974. The Roots of Coincidence. London: Hutchison and Co. Mackie, J. L.1976. Problems from Locke. Oxford: Clarendon Press. Minkus, H. 1977. “Causal Theory in Akwapim Akan Philosophy”. In Wright, R (ed). African Philosophy: An Introduction. Washington D. C.: University Press of America. Mosley, A. 1978. “The Metaphysics of Magic: Practical and Philosophical Implications”, In Second Order, Vol. 2. Nos 1 & 2. Ile-Ife: Ife University Press. Mundle, C. W. K. 1967. “ESP Phenomena, the Philosophical Implications of”. in P. Edwards (ed.), The Encyclopedia of Philosophy, Vol. 3. London: Macmillan. Nana Ofosu (a diviner-priest, at Amoakrom, Ghana). Interview with, on the 5th of September, 1999. Oluwole, S. 1995. “On the Existence of Witches”, In Mosley, A., (ed.), African Philosophy, New Jersey, Prentice-Hall, p. 366. Parkinson, G.H.R.1965. Logic and Reality in Leibniz’s Metaphysics, Oxford, Clarendon Press, p. 56-75. Powers, J.1982. Philosophy and the New Physics. London: Methuen. Ramose, M.B. 1999. African Philosophy through Ubuntu. Harare: Mond Books. Rattray, R. S. 1927. Religion and Art in Ashanti. Oxford: Oxford University Press. Rhine, J.B. 1934. Extra-Sensory Perception. London: Faber and Faber. Russell, B. 1937. A Critical Exposition of the Philosophy of Leibniz. London: George Allen & Unwin. Schmidt, H. 1969. “Precognition of a Quantum Process”. In Journal of Parapsychology, Vol. 33. 99-108. Sklar, L. 1992. Philosophy of Physics. Oxford: Oxford University Press. Sogolo, G. 1993. Foundations of African Philosophy. Ibadan: Ibadan University Press. Martin Ajei 43 Stanford, R. 1978. “An Experimentally Testable Model for Spontaneous Extra- Sensory Events”, In M. Ebon (ed.) The Signet Handbook of Parapsychology. N.Y: American Library. Tyrell, G. N. M. 1953. Apparitions. London: Gerald Duckworth and Co. Utts, J. 1991. Replication and Meta-Analysis in Parapsychology. In Statistical Science. Vol. 6 No. 4. 363-403. 45 SUBJECTS, AGENTS, OR COLLECTIVES? THE DISCOURSE OF YOUTH AND PHILOSOPHY Ibrahim Bello-Kano1 To become what one is, one must not have the faintest idea what one is —Friedrich Nietzsche. Ecce Homo, p. 254. Abstract The present paper argues that the term “youth”, which is traditionally used to refer both to young people of a certain age bracket and to a time of life between childhood and maturity, has acquired distinctive yet contradictory meanings since the 19th century, and that the category of people, individuals, or persons that the concept describes or purports to analyze (the so-called young people, teenagers, pubescents, adolescents) may be regarded as subjects in the philosophical sense of being persons capable of intentional behaviour and to whom intentional predicates (beliefs and desires) can be ascribed but not, however, as a collective agent, with the capacity for goal-directed activity (such as, for example, political, social, or national transformation), in spite of the shift in the use of the concept from a singular to a collective noun. The paper argues further that the term “youth” is a vacuous concept, and thus lacks any philosophic or analytic significance or explanatory value in social theory and, especially, in philosophy, and that the discourse of youth which deploys the concept can only sustain the “politics of collective singularity” whereby a singular or a single collective subject or a parasitic structure usurps, or feeds on, the activity and capacity of empirical subjects (young people). The paper draws out the philosophic and practical-political implications of its central arguments— namely that young people, teenagers, pubescents, or adolescents those presumably described by the collective noun, “youth”, do not, and cannot, articulate a coherent group - or age-based beliefs, desires, reasons, and action; cannot represent (or be the collective agency of) definite, historically-specific political-economic interests or relations in society; and, that, to the same extent, cannot be an agency of, or for, and indeed cannot be mobilized for, any form of enduring political action or social or national transformation. 1 Professor Ibrahim Bello-Kano is a Professor in the Department of English and French, Bayero University, Kano. 46 Legon Journal of the HUMANITIES Volume 20 (2009) 1. Introduction That there is something, some transcendentally real presence, some group of entities, collectively called youth is widely believed. Whether as term or concept, youth purports to describe a specific set of human subjects (adolescents, teenagers, and young people in general), a process (a time of life that is neither childhood nor adulthood, that is, somewhere in-between), and a transition (the period between childhood and maturity). Yet the actual age bracket for youth is notoriously variable: for the United Nations and the World Bank, for example, it is 15-24 years, while the Commonwealth Youth Programme specifies 15-29 years; and for many countries the figure varies from 13-18, 20-25, and even 12-20 years. The Nigerian National Council on Youth Development defines youth as “all young males and females aged 18-30 which are citizens of the Federal Republic of Nigeria” (National Youth Policy of Nigeria, p. 4)1. What is more, virtually every country and multilateral institution has got a “Youth Policy” and a “Youth Action Plan”, as can be seen in the case of the Nigerian National Youth Policy, The ECOWAS Commission Youth Policy, African Youth Charter, Commonwealth Youth Charter, United Nations World Programme of Action for Youth, and so on. In 2006, the Obasanjo Administration even established the Ministry of Youth Development. The Nigerian Youth Policy, for example, takes as axiomatic the view that youth is a natural kind (young people, the young), what it calls “the greatest assets that any nation can have… the future leaders… the greatest investment for a country’s development… a vital resource” (p. 1). These statements about youth, from those of the Nigerian National Youth Policy to the United Nations World Programme of Action for Youth, all assume that young people exist (as natural or naturalized entities, assets, and resources), and that the word or term “youth” describes a concept, that of entities, assets, resources, and human organisms (young people of a certain age-bracket). Indeed all of those institutions conceptualize young people, or youth, as the “future” (in terms of human biological and social reproduction) of human civilization, not just of specific countries, communities, or societies. All speak of “empowering” young people in terms of political, social, economic, and development efforts; all pledge to put youth at the centre of national, regional, and international development. It is thus clear that some notion, idea, or concept of youth is discernible in the discursive practices of many institutions, from the state, the family, and society to the market economy, whether it is a government department concerned with “Youth Development”, or “Youth Employment”, the legal system or penal code concerned with “Youth Crime”, a traditional family anxiety about “youth behaviour”, or “transition to adulthood”, or market advertising concerned with “Youth Culture”. However, it will be shown below that the term youth is, and Ibrahim Bello-Kano 47 describes, a more or less contradictory and unstable set of entities or natural kinds (teenagers, adolescents, young men and women, “future leaders”, human resources, the young, etc.), and that these entities are, in the end, not meaningful things in themselves, not units or entities given in nature, but an absence, a difference (a differential network), rather than a reference (a presence). 2. Analytical Framework This paper seeks to put epistemological and ontological pressure on the concept of youth (as a collective noun) in order to determine how far it can go, and how much it can do, paradigmatically and causally. The paper aspires to be a philosophical critique concerned with the adequacy of concepts, the consistency (and rationality) of propositions, and the validity of arguments. This form of critique is a kind of explanatory critique which aims, in general, to identify the presence of causally significant absences in conceptual schemes, what cannot be articulated or done in a specific language or conceptual scheme, and what is said or done or presupposed by means of such language or system. This form of critique may be called “meta-critique” in that it strives to unite the criticism of concepts and their conditions of possibility; it illuminates or explains conceptual errors by seeking to work towards the elimination of the condition, action, or practice that informs, conditions, sustains, or necessitates those conceptual errors. Here, the critique passes immediately and unconditionally to a negative evaluation or judgement of the objects (ideologies, politics, social relations, etc.) which make such conceptually or cognitively defective conceptual forms necessary.2 Thus on the terms of meta-critique, while the meaning of the word “youth” (as a collective noun) actually depends, as we have seen, on reference to a set of circumscribed entities (young people, for example), it is nevertheless the repeatability of the word (what is meant by it by those who use the word as a descriptive or analytical term or concept) is not something that attaches in a causal relationship to meaningful things or strings of things (to entities given in nature) but to a network of traces (that is, the differences in the entities held to be true of youth, namely assets, people, resources, the future, potentialities, etc.). To use Saussure’s opt-cited phrase, “in language there are only differences without positive terms [reference]” (1974: 121). Now this suggests that it is the repeatability (the meaning) of the word “youth” that is a condition of its being a socially meaningful word; for only by being part of a chain of possible repeatings can the word have meaning at all. It is this that establishes the field of difference (the need for the word to describe more than just young people but their attributes, and other classes of things, as we have seen) that confers on the word its meaning (either as a sign, a signifier, or as a conceptual signified). That is, the activity of 48 Legon Journal of the HUMANITIES Volume 20 (2009) conferring meaning on the word “youth” (or what it purports to describe) owes its intelligibility not to a reference (a presence; young people as empirical subjects), a sameness (all young people) or determinacy (the biological, physiological, social, or cultural irreducibility of young people, for example). The wider point is this: there can be no movement from (word, concept, term) to the real thing (some immediate presence, some unity called “young people” or “adolescents”, to which the “real thing” now stands). A concrete illustration of this is the differential ways in which the word “youth” has figured in “statements” (“policy”) about young people or what we may call “the discourse of youth”, where, indeed, we find no relation of reference that anchors the word “youth” to things or entities in the world, or what, following Derrida (1976) we would call the structural relation to a presence that is always elsewhere. 3. Discourses of Youth As is usual with concepts and descriptive terms, the word “youth”, which is variously deployed with reference to a distinct category of individuals, to the time of life between childhood and maturity, or to a person within that time of life, has had shifting and mutable uses, and has acquired, even in the academic literature, distinct yet contradictory meanings, at least since the 19th century (Gillis 1974; Fritz 2005). This section examines in detail the specific conceptualizations of the word “youth”, and the specific discourses (meta-statements) that “objectify” the word, and the concept of youth as an empirical subject. The term “youth” elides three overlapping situations. First, the biological- physiological aspects of the age-process, variously described as “puberty” and “adolescence”, which covers persons in the 13-19 years bracket3. But in this sense, “youth” (rather than puberty, for example) describes a sociological situation, rather than the physical-biological changes involved in human development (Fritz 1985). Second, the transitional stage between childhood and adulthood, described by the term “young man”. But in this case, “youth” describes a “stage of incompletion”, a transitional process, rather than the vexatious social and sub- cultural spaces which young people are said to be likely to occupy due, in part, to what Aristotle would regard as their lack of experience of particulars. Third, is a situation described by Fritz (2005: 380) as the “rebalancing of youth’s positive and negative connotation”; that is, a re-description of the negative connotation (i.e. ignorance and recklessness of a youth) and the positive ones (the flexibility, vitality, and freedom of young people) to mean that youth is not only a specific determinacy (biological-physiological-psychological changes, objective time, definite patterns, and limitations) but also indeterminacy and incompletion (a Ibrahim Bello-Kano 49 movement towards complexity and valuable complication, a transition to a higher order, etc.). In this sense, youth no longer marks the transition from childhood to adulthood, or even a specific or actual age (e.g. “teenager”, “childhood” or “puberty” or “adolescence”) but a free-play of style, a free self-definition, a life- style, or, in fact, the social institution of self-creation and self-making4. It was partly in response to those situations the term “youth” had to change, and did change, from a singular noun (e.g. “a youth”) to a collective noun (“youth”), which now describes not only specific persons or individuals within a particular stage or time of life but also a state of mind, a life style, so that even adults and older people (in their mid-life or after) can be young as a matter of life- style or feel young as a matter of personal choice. What is significant about all this is that it is now no longer possible to distinguish clearly and decisively those shifting and contradictory meanings of the term “youth” in either the public understanding of the term or its analytical implications in the academic literature. For example, Gillis (1974) has argued that many urban societies across the world saw the transition from childhood to adulthood as fractious and problematic; and that this awareness intensified as more and more families moved from rural to urban areas, and as the permanent revolution of modernity and modernization imposed massive pressures on personal identity and traditional modes of living. And with the intensification of the facts of spatial and technological urbanization, with the development of industrialization, bureaucracy, professional work, the market, universal education, and time- consciousness, all of which combined to disrupt the normative and socializing function of the family and other traditional units, youth came to describe social dislocation and problematic identity, crime, urban alienation, and undesirable behaviour such as drug addiction and sexual hedonism (which today includes anxiety about youth and HIV/AIDS infection). The usual response from society and the state was to see “youth” as a social problem which required resolution within specific institutions such as “Youth Courts”, “Youth Prisons” (also called Juvenile Courts or Juvenile Prisons) and “Youth Services” (for correctional and counselling measures within the wider institutions of social control). All of this means, of course, that the term “youth”, implied, in those contexts at least, the possibility of what Fritz (2005:381) calls “a permanent state of irresponsibility” (as may be illustrated by the Nigerian situation of the reckless driving by commercial motorcyclists who are mostly young rural migrants). Nevertheless, within this institutional response to “youth”, there also exists, and has existed, a positive affirmation of youthful activity or youth self- making. This is conveyed by such terms as “youth clubs”, “youth workers”, “youth opportunities”, “Young Farmers”, “Young Turks”, or “Young Conservatives”, 50 Legon Journal of the HUMANITIES Volume 20 (2009) and in Nigeria by the National Youth Service). All of this collectively affirms the universal image of youth as a desirable, transcendentally benign collective existence. Yet even here, “youth” describes a masculine category, in two contradictory ways, namely a male-working class (which excludes women), with associations of gender-based leisure and work spaces and class resistance. Here youth is signified by their male youth sub-culture, and by “their deviant and often bizarre style of dress a way of making sense of their marginal position in society” (Billington et al 1991: 132. See also, on this, Hall & Jafferson 1976; Mungham and Person 1976; Hebdige, 1979). Another is the positive image of youth as a form of desirable consumption and an attractive subjectivity, a normality hinged on market-exchange, and in young persons’ transition to adult responsibility through work and cultural exchange. According to Fass (1977), it was the public anxiety about consumption and (desirable) market-exchange relations, most specifically in concerns about “hedonism” and unbridled consumption, that a new conception of youth emerged in the late 20th century. This signalled a further shift, in many affluent urban cultures from a conception of youth as a problem of order (which the state should tackle through institutional mediation) to one of youth as the epitome of existential autonomy, self-indulgence, pleasure, and desirable physical qualities such as health, beauty, vitality, freshness, and market choice in terms of consumer goods such as cigarette, designer cars, clothes, and shoes. This is what Fritz (2005: 381) calls youth as “a market position”. In this sense, youth has become a symbolic and aesthetic position, a life-style. It describes, and relates to, the modern, even “post-modern”, self-fashioning, and the aesthetics of identity nurtured by consumer capitalism, market-advertising fantasy, and, especially, Western forms of consumerist chic, for example, obsession with youthfulness, youth culture, youthful body, sexual attractiveness, sexual virility, health, vitality, physical beauty and public glamour — in dress, speech, gesture, and personal appearance in general. This notion of youth has become, despite continuing state interest in youth as a public or social policy (youth crime and delinquency, youth unemployment, etc.), the dominant late 20th and early 21st century conception of youth (as a collective noun). Consumer capitalism has re-defined and re-described youth to mean “consumers”— the target for fashion clothes, cosmetics, computer games, and sports merchandise. Advertisements of virtually everything the market produces (household equipment, cars, clothes, food, drinks, cigarettes) now promote the idea that all consumers are, or should by implication be, young people, or youth as such — upwardly mobile, fashionable, glamorous, and responsible (i.e. able and willing to be all those desirable life-long consumer identifications). Ibrahim Bello-Kano 51 It may in fact be argued that youth as a model of consumption (rather than as a problem of order or a problem for social policy) has become the new ideal of subjectivity in late modernity. This is the case because virtually all the actually existing youth policies we have seen, from the Nigerian Youth Policy to that of the United Nations, stress, as their guiding principles, goals, and objectives, the integration of youth into national reconstruction and development. The Nigerian Youth Policy, for example, seeks to, in its own words, “put the development and participation of young people at the centre of national development efforts ... to mobilize their potentialities … take charge of their own destiny [and become] active participants in the shaping of the political and economic destiny of [their] nation” (pp. 1-2). From Nigerian Youth Policy to that of the United Nations, this imperative of youth involvement is no other than the rhetoric of the many facts and faces of the neo-liberal philosophy of capitalist development through the twin programmes of free market and liberal democracy, and their accompanying rhetoric of human rights. It is thus fair to conclude, on the basis of the preceding discussion, that youth in the present circumstances describes contradictions (in the conceptualization of young people) rather than the problem of transition (within the time of life) of young people. But before we examine this issue in detail, let us consider the problem of whether the word “youth” does indeed capture the metaphysical complexity of young people. Now young people exist in all societies, despite the fact that people are continuously being born. Do young people, it may be asked, view, conceptualize, or identify, themselves as young people, or, in this case, as youth? Young people do indeed form a sort of empirical subject, that is, both as human organisms and as persons. The question is whether their being persons indicates their metaphysical integrity as a collective, which the collective noun ”youth” describes or identifies. 4. Intentional Predicates: Personhood The preceding discussion indicates that the items we call youth, or that are designated as youth in our culture, only exist in what we have referred to as a chain of possible repeatings within language and meaning. That is, as the multivalent uses of the word or concept of “youth” have shown, the word youth acquires its sense or meaning by being “repeated” at many levels. And as has been shown in the preceding pages, the word has never had, or in fact never acquired, a semantic determinacy, or a sameness of meaning; for as we have seen in the case of the uses, senses, and meanings (significations) of the notion of youth, there have not been, neither are there existing, a set or sets of ideal semantic entities from which to derive this sameness of meaning (i.e. the ontological identity of 52 Legon Journal of the HUMANITIES Volume 20 (2009) youth as a structure of reference to a real thing, namely young people). The question, then, is whether youth does have the metaphysical depth or personal agency of real people; and whether, indeed, youth is a determinate presence, with a secure metaphysical foundation and integrity. In other words, is youth self-posting? We can answer this question at two levels, that of “intentional behaviour” and of “subjectivity and agency”. Any entity can be regarded as a person, with the capacity of personhood, if, according to Dennett (1981: 269-71; 281-85) it can meet the following conditions: Ratii. onality (attitudinal and behavioural) Intentionality (intentional predicates such as beliefs and desires can be ii. ascribed to it) The iii. attitude taken towards it (the stance adopted with respect to it) It is capable of iv. reciprocating in some way It is capable of v. verbal communication (dependent on possession of language) It is vi. conscious in some special way (i.e. it is alive, not dead, and is aware of having engaged in actions for which it can be held responsible). The first difficulty we should raise is whether we can ascribe attitudinal and behavioural rationality to the collective noun, youth. We can always ascribe beliefs and desires (intentionality) to persons or people because we can assign a sense to their utterances or language-use and to their observable behaviour (which may typically admit of more than one interpretation). The problem is how to know which beliefs and desires to ascribe to persons. Dennett (1981:19) proposes that “we get round the ‘privacy’ of beliefs and desires by recognizing that in general anyone’s beliefs and desires must be those he ‘ought to have’ given the circumstances”. But this typically requires the person having some causal and ontological irreducibility since it also requires that the person is capable of changing her beliefs to ensure, for example, that her beliefs are true, at least in a minimal sense (attitudinal rationality); and that the person is also capable of acting in the light of beliefs and desires (behavioural rationality). Only when this is the case can a person be said to be rational, and capable of “reflexive monitoring”, her being aware of her “own-states of awareness” during her activity; that is, her capacity to monitor the monitoring of her activity. Now it can easily be seen that youth is not, in the senses established in the preceding argument, a person. Since the term (as a concept and as a sign of presence) gathers and groups together amorphous, discrepant, and mutually Ibrahim Bello-Kano 53 exclusive “semantic” strings (a chain of possible repeatings; a chain of signifiers), it lacks the capacity, even the possibility, of a reflexive monitoring of its own intervention in the material or social world. It is even more difficult to ascribe intentional predicates to youth, since it is not intentional under such analytical terms as retrospection (retrospective commentary upon its actual or imagined activity in the world). As Bhaskar (1979: 104) would argue, “any entity X that lacked the capacity to refer to its own states of consciousness (and to interiorize references to itself in the third person) could not use these states of consciousness for the production and communication of information”. This is the rational significance of Dennett’s conditions of personhood, namely intentionality, reciprocity, language, and consciousness; for these not only specify the necessary conditions of personhood (agency) but also the necessary conditions for any discursive (non-intuitive) intelligence. To use a formulation of Bhaskar’s, the facts of both “intervention” and “commentary” are “always the situated doings of agents at places in time” (1979: 104). But this is, paradigmatically and structurally, denied to youth as a transcendental entity; for as a collective noun, it lacks intentionality, reciprocity, reflexivity, language, and consciousness — for these are the attributes of persons, or what we may call “individual human subjects”. This is the case because only individual human organisms have any metaphysical integrity in the light of the conditions of personhood specified above. The upshot of the argument is that any human person has, or possesses, this metaphysical integrity, which, in substance, amounts to capacity: for any person as person shares this capacity, irrespective of age, ethnicity, gender, sex, life-style, disposition, ideology, beliefs, identity, or her place in society, or in the relations of production (class). In other words, youth as such does not possess some separate set of intentional predicates, or intentionality, from any objectively-given human organism. Thus on the basis of what is called the Principle of Humanity5, individual young people could be persons but not youth (as a transcendental entity or as a collective noun, each of which implies the existence of a collective subject which, if we would only uncover or identify its presence would fix its “authorial” meaning for us). 5. Subjects, Agents, Collectivities Individual human beings may be defined in terms of their tendencies and powers — as agents who act on the basis of beliefs and reasons, which are themselves tendencies in the real world. Now the question is whether youth has an identifiable action on the basis of its position, interest, beliefs, and reasons. To ask the question in another way: is youth an agent (a causally efficacious subject)? First of all, only human beings, as intentional beings, and as beings 54 Legon Journal of the HUMANITIES Volume 20 (2009) to whom intentional predicates can be ascribed, are subjects. For intentional predicates, rationality and language, make human beings subjects. On this view, “the subject is what speaks” (Belsey 2005: 40). This is why for subjects, social action is ultimately connected to the subject’s speech (given the centrality of intentional predicates in specifying personhood and, by extension, agency). On the basis of this insight, we should have no difficulty identifying agency at this level — the subject’s verbal behaviour, which can, excepting deeper philosophical difficulties in terms of inferring meaning without belief6, be straightforwardly read off any evidence for it (in the form of authorship and linguistic and discursive responsibility). Despite arguments on the basis of the so-called Principle of Charity, according to which we can take “the fact that speakers of a language hold a sentence to be true (under observed circumstances) as prima facie evidence that the sentence is true under those circumstances” (Davidson 1984: 148), youth has no speech or language or sentences or a fabric of sentences that could specify its agency status; for youth has no utterances that we could interpret as indicating its beliefs (and reasons). This is why we cannot have any reason to count youth (either as creature or presence) as rational and as having beliefs, or, to borrow a metaphor of Davidson’s, as saying anything. For example, note that youth as agency does not speak, and is silent in the discourses of Youth Policy, from the Nigerian to the United Nations’. Someone else speaks for youth in these discourses. Thus on this minimal sense at least, we cannot say that youth is an agent since we cannot interpret its speech, utterances, or verbal communication chiefly because interpretation involves ascribing beliefs (and reasons, which are themselves beliefs, and which may themselves be true or false) to agents (and who, by necessity, are attitudinally and behaviourally rational). Thus, unless we assume that “agents are rational, their doings and sayings are unintelligible” (Callinicos 1989: 114). This is the transcendental argument (on the basis of specific individual agency) against the assumption in the discourses of youth and national and international youth policies and programmes that youth is an agent, and possesses the attributes of agency (even on the basis of individual agency to which beliefs and desires can be ascribed on the basis of the Principle of Humanity, to which Dennett’s six-point condition tacitly refers). Nevertheless, as we have seen above, youth is sometimes seen as describing the structural and organizational capacities of young people (which is signalled by the many “youth wings” of political parties, the youth cadres of many, especially rural, communities, the UN, ECOWAS Youth Charter, and the Nigerian Youth Policy, etc). In these discourses and their operational organs, youth is re- described as a collective agent (or agency). But what happens if youth is, as argued previously, not an individual agent; is it, then, or can it be, a collective agent? Ibrahim Bello-Kano 55 This question turns upon the structural-organizational capacities of collective agents, their consciousness of themselves, and their sense of common identity (as a collectivity). Individual agents may form a collectivity in order to pursue their desires, reasons, interests, and objectives. Typically, a collectivity seeks to achieve some goal or goals, pursue its interests, and seeks to remold society in the light of those goals, interests, or objectives. Thus a “collective exists where persons co-ordinate their actions because they believe themselves to have a common identity” (Callinicos 1989: 135). Such identity may be, again, a set of predicates designating properties unique or special to such individuals, namely class position, or some position in productive relations, or the social relations of society (which could indicate social power of sorts, for example, ethnic organization, political or administrative position in the state apparatus, etc). As Callinicos argues, unless individuals “believe themselves to have something in common and treat this as the basis of their collective action, they are not a collectivity” (135). We may say, then, that the unicity of young people as subjects and agents is not present in the concept of youth. So in what sense, then, can we say, for example, that youth forms, or describes, a class position (or a definite identical position in the production relations)? But note, following Giddens (1981: 111-13), that while class awareness does not have to acknowledge the existence of class position, class consciousness does have to. The wider point is that it is class consciousness alone in the form of what he calls class identity, conflict consciousness, and revolutionary class- consciousness or “the recognition of the possibility of an overall reorganization in the institutional medication of power … and a belief that such a reorganization can be brought out through class action” (111-13), which defines class as a collectivity. This is not to correlate class consciousness with the actual existence of a class but merely to suggest that the collective action of individual agents, the collectivity, is a necessary condition for class struggle. But this is precisely what youth lacks: young people qua young people do not share identical class positions, and for this reason cannot, and do not normally, have a collective consciousness (a “collective youth consciousness”). Young people qua young people have no sense whatsoever of the “recognition of the possibility of an overall reorganization in the institutional mediation of power”; or a belief that “such a reorganization can be brought about through class action”; or even collective youth action, on the basis of a “collective youth consciousness”. (But even if individual “youth” agents could imagine this, it would not be a realistic possibility, but, perhaps, only a negative utopia)7. And in a situation where youth describes an amorphous collection of individuals without a common language, common interests, beliefs, reasons, and culture; and scattered across ethnic, gender, class, religious, communal and 56 Legon Journal of the HUMANITIES Volume 20 (2009) national identifications, it cannot be said to hold the promise of, or prospects for, what Mann (1986: 219) calls “the capacity for extensive organization” (as, say, the working class or the bourgeoisie is capable of). Thus on the basis of this insight, the following implications are evident: youth is 1. structurally incapable of organization by reason of its own collective interests youth has no identifiable 2. structural capacities whatsoever youth is incapable of 3. social action (as a collective) youth is incapable of 4. social consciousness (a universal “collective youth consciousness”) youth is 5. discursively incapable of accessing its own “collective” awareness youth is incapable of what Mann calls “intensive power”, the ability for 6. tight organization and command based on a very high level of collective mobilization and commitment (1986:7). Moreover, without the possibility of “value-consensus”, internal consistency in values, a sense of an alternative to the present, a vision of collective action as identity; without a sense of itself as a conscious actor in the world, a modern subjectivity in its own right, autonomous and sure of itself; without any sense of its social identity as a subject; without, in the end, any sense of a “practical consciousness”, youth is, and remains, a metaphysical fiction (since youth has no collective interests or positions that it is compelled, paradigmatically, structurally, and causally, to defend). Even more important: youth lacks any ontological integrity (since it is not even a social group, what, borrowing an analogy of Gramsci’s (1971: 333) we might call the “active people-in-the-mass”). Rather, youth is, at best, a patient in the discourses of the state and other institutional youth programmes, a “raw material” in the discourses of Family-Adult institutions; and, is properly speaking, no more than a “statement”, a discourse of heterogeneous institutional-cultural elements. 6. Productions of Discourse We have seen above how youth is the production of not only the signifier “youth”, specifically the biological-physiological aspect of the age-process (“puberty”), the transitional stage between childhood and adulthood (“young man”), a state of incompletion in social and cultural space, a life style within the social institutions of Romantic self-creation, a state of mind, a problem of order (a social dislocation and problematic identity requiring public or social-institutional policy), and as a market position (youth as an ideal of consumption). Now all Ibrahim Bello-Kano 57 those “positions” may have their intrinsic language (or codes), or a symbolic and symbolizing system, which “fixes” the meaning of the so-called entities they over- determine. The way this “language” or “symbolizing mechanism” divides up young men and women as empirical subjects, for example, is, and becomes, the meaning of youth itself (that is, by displacing youth as a presence). But because meaning itself is differential without a positive content, or a structure of reference to which the word, term, or concept “youth” may be compared in an affirmative sense, the concept of youth becomes a sign, a signifier, whose form and meaning (signified) are another sign (another word); or this “meaning” only derives its meaning from its role in sentences (discourses), the relations of sentences (discourses), without yielding the real thing, so that the “truth” of the meaning, if there is one, lies elsewhere. This is the significance of Derrida’s bon mot that the thing, the object, itself is a sign). This is precisely what we have seen in the case of the meaning and form (“use”) of the word/concept/term “youth”: the sign/signifier “youth” exists, and is apparently used and appropriated by heterogeneous discourses and institutions, without reference to the transcendental pre-propositions of experience. Yet as the work of the analytic philosopher, Donald Davidson, has shown, there is no relation of reference that breaks us free from the holistic connections within language and anchors individual words to things in the world. There are no propositions like entities that language mirrors, and we cannot steer individual sentences into confrontations with the world that make them, individually, true or false. There are no meanings a grasp of which allows a translator to come to rest at finally having brought the other language into contact with what is really meant (Farrell 1996: 256-57. See also Davidson 1984: 73-74; 215-26). The implication of all this is that the concept of youth does not stand for pre-existing entities (or even pre-existing concepts) but only for a system of concepts and forms which organizes the discourses of youth, without ever breaking free of this system of difference, and onto an experiential structure of reference, onto the empirical entities called “young people”. Hence we are left with no alternative than to see the concept of youth as an interpretation, indeed the interpretation of an interpretation, and, in Derridean terms, the trace of a trace; for both interpretation and the trace have, to borrow a phrase of Belsey’s, “no final guarantees elsewhere” (2002: 22). In a word, both 58 Legon Journal of the HUMANITIES Volume 20 (2009) the concept and the discourse of youth are rather the production of the free play of the signifier, a differential network of traces8, for as Derrida writes (1976:57), “difference cannot be thought without the trace”; that is, they are traces of other traces, in an endless chain that never yields the transcendental signified (the subjectivity and agency of youth). Which is to say, in other words, that youth (both as word, differential concept, and signifier) is a citation, a structure of citationality (alluding to something but with what is alluded to being crucially different from the allusion itself). Another way of conveying this is to say that there is really nothing “behind” the citation; it is nothing but the space on which all the “quotations” and “citations” which make up the “discourse of youth” are inscribed (as substitute significations)9. Nevertheless, we cannot say that youth is wholly and completely the product of language. It is also the product of material and social institutions (or what may be regarded as material practices and structures), or what, following Foucault (1972) we could call “discursive formations”, which, he argues, are not only made up in languages in use (“statements”; “a system of dispersion”) but also of “heterogeneous elements”. He calls these “practices that systematically form the objects of which they speak” (1972: 49). According to Foucault, human subjects are produced by discourse or discursive (social) practices, such as for example, investigations, talk, and writing by doctors, clergy, novelists, moralists, and politicians. These discursive formations actually bring about the “thing”, say madness, sex, etc.,10 they claim to have discovered. It is in this way, for example, that the “thing” is constructed by the discourses of social practices and institutions. Such discourses give the impression that “the thing” has existed prior both to the discourses and the discovery of it. For Foucault, the reverse is the case: these discursive practices have in fact constructed “the thing”, the object (of these discursive practices), in order to analyze, describe and regulate (control) the activities of human beings. This control takes many forms, one of which is that these discursive practices are repeated across very many and different social- discursive fields; and they go on from there to “authorize” (produce) a certain speech position, construct certain objects and truths, which they would then endow with a certain reality, with a certain locutionary determinacy11. This issue connects with Foucault’s other central concept, that of “power- knowledge”, what he calls the “relations of domination” that articulate discursive practices into a historically specific apparatus, for example, the disciplines, the practices of surveillance and control, such institutions as prisons, schools, asylum, factories, and sexuality (1980: 194-95). For Foucault, power is a key element in discourse in that it is these practices that constitute subjects, for, as he writes, “certain bodies, certain gestures, certain discourses, certain desires, came to Ibrahim Bello-Kano 59 be identified and constituted as individual”. Therefore far from power being a repressive apparatus which is imposed outwardly on the subject, the individual, the subject, is, rather, Foucault argues, “ not the vis-à-vis of power [but] one of its effects” (1980: 98; my emphasis). One implication of this for our discussion is that “youth”, even where the concept is used with a referential intent, is rather the production of discourse, or, indeed, of those “discursive practices” (discursive formations) we have identified above, from biological-physiological, psychological, familial, sociological, moral, historical, advertising, and masculine discourses. It is these discursive practices that have brought youth (both as a collective and as a collective noun) into being; and it is these discourses that have sought to “recruit” youth as a subject subjected to their goals, values, and interests; it is these discourses that have disciplined youth into a “meaningful” concept (i.e. as anxiety, chaos, boundless energy, crime, market, life-style, vital resource, asset, and subject-position, and whose sole relevance is the advancement of national and international market exchange, political-economic development, and modernization); and finally, it is these discourses that have formed “the knowledge” with and within which youth becomes the product of the subjugation of objects (“youth bodies”). Which is to say, to follow an analogy from Foucault, that the discourse of youth not only imposes “canonical bits of knowledge” on young people (as youth) but also uses them as a basis for construction a “science”, a normativity, that is beyond their grasp. All of this together enables the discourse of youth to be, in the words of Foucault, an “intensification of the interventions of power to a multiplication of discourse”; for the very multiplication and the multiplicity of the discourse of youth is, in the last analysis, an “intensification of the interventions of power” in the life-knowledge-being of young people, so that even young people themselves are subjugated to an object called “youth”, that is, are “repeated, renewed, and displaced” (Foucault 1981: 70; my emphasis). If this is indeed the case, then youth is not a consciously willing subject creating its youthness (youthfulness or being-in-itself) from a point outside discourse; it is not even a self, or a demiurgic subject that has a stance outside discourse, but rather as a localized position with it. To this extent, then, the use of the concept of youth by the state, the market, sociology, and social policy, for example, are decisively and irremediably discursive: youth is a discursive construct, an effect of cultural production and symbolic representation, not a determinate meaning, or even a subject – position (or a subjective activity). 60 Legon Journal of the HUMANITIES Volume 20 (2009) 7. The Logic and Politics of Substitution The discourse of youth, the notion that youth is, or could be, a subject of history, an agent in national development and political participation or transformation, the idea that youth is transcendentally real and efficacious, or that it is a collective singular subject, is the typical dream of presence, the myth of the given, in which reality is directly given to consciousness or to the subject (parousia). In the discourse of youth, as I have been using the expression, youth is present to itself: it guarantees the discourses of it, statements about it, and supplies metaphysical sufficiency and guarantee to the idea that youth is a person, a subject of proposition and action, and the sign of potential or actual collective action (national development, youth development, democracy, and political participation). Now what all these notions foreground is a model of youth as a being which is capable of matching its desires and intentions with its potential (future leadership, vanguard of national development, progress, etc.). This also means that youth is a single, collective subject, a singular entity defined by its (collective) singularity. How this so-called collective singular subject is formed materially, culturally, linguistically, and discursively is never specified. What is more, such a notion of youth does not specify whether its “goal” is social production or only instrumental action in league with its nature and purposes. In any case, any model, theory, or discourse which privileges one social group as the bearer of society or humanity or assumes the existence of an epistemologically transparent structure of self or selves which possesses transparent knowledge for determining what is or will be, or what would actualize its self-becoming, is a form of the “philosophy of the subject”12. One philosophic difficulty of this notion is that it is, as Benhabib (1986:135) has pointed out, based on a pre-linguistic model, which proceeds from “a reflecting consciousness formulating its intentions and goals”. This reductionist mode of thinking obscures the extent to which human beings attain their dignity as persons wholly within the “shared social world” and through “linguistically mediated socialization”, not to speak of the cultural-symbolic codes which make human beings subjects. And so by a miraculous process, the discourse of youth as a collective singularity has turned its object into a pre-linguistic, pre-social, and demiurgic subject (namely that what it is and what it can accomplish in its activity are fundamentally independent of all else). The flip side of this “messianic” (instrumentalist) notion of youth is the politics of collective singularity, a “mode of politics where one group or organization acts in the name of the whole” (Benhabib 1986: 347). The obvious implications of this are an authoritarian politics in which a singularity or a collective Ibrahim Bello-Kano 61 singularity acts in the name of the whole and imposes the normative standards of conduct or behaviour (which, in their turn, can foster an attitude of dependence and clientalism). Yet the same logic and politics are operative in the notion of youth as an empirical subject, or a social-economic patient, needing restitution or reconstruction by state institutions such as the National Youth Policy, the Ministry of Youth Development, Youth Directorates, the National Youth Service Corps, Youth Wings (within political parties), and youth studies in the academia. All this proceeds from the assumption that young people (or youth) possess a normativity in their empirical manifestation. Quite apart from questions of the inherent interpretive indeterminacy of action (given the interpretive indeterminacy of the life-world), it is philosophically untenable to privilege the particular consciousness, perspectival knowledge, or identity of a group or collectivity, first because consciousness is perspectival (agents know the world through a human perspective, and make distinctions based on the symbolizing systems of socialization), second because knowledge is propositional (which makes immediate knowledge impossible, since there is no Archimedean point outside discourse), third because experience is always already conceptualized (which makes perception subject to multiple interpretations), and finally because meaning is immanent (in language, which implies that meaning cannot be anchored to extra-linguistic mechanisms such as physiology or psychology). The implication of this for the discourse of youth is obvious: there is no identity of subject and action because one cannot predict in advance what a subject or an agent is obliged to do in the light of its subjectivity or “agent-ness” (agency). And since no epistemologically transparent selves (can) exist, not to speak of self-actualizing collectivities or singular collective subjects, then it is a philosophical mistake to assume that there exists a self-knowing organism or collective singularity which can act, or transform society, solely on the basis of its non-signified self-presence or self-presentness (or self-origination) in relation to an action, a goal, or (inherent) purposes. Even more important: to claim that a group of persons, youth in this case, exist as empirical givens, and as those who have actualized themselves through their age or quality (youthfulness, of being young) is to claim that their potential, essence, and faculties are identical to their being, existence, action, and doing. This, as said previously, leads to the politics of collective singularity, whereby a singular or a single collective subject acts on behalf of others, or in the name of the whole (humanity, young people, workers, peasants, an ethnic group, etc). There is another name for this: messianic collective singularity, for which there is a man, a group, a collective, within human society whose strategic position or being or 62 Legon Journal of the HUMANITIES Volume 20 (2009) identity or experience uniquely entitles it to represent the whole of society or the plurality. This, surely, is the secret of all the different kinds of the discourse of youth that we have examined: they construct young people as patients, namely they are subjected to monetized, bureaucratic, instrumental, administrative, and policy-decisions norms of action. All this would only, in the end, impoverish the conditions of young people and foster in them an attitude of dependence, which would also only further limit, and not enhance, their capacity for autonomous and self-creative action. Not only would such instrumentalist conceptions of young people seek to reach decisions at their expense (in the long run), not only would such limit their normative processes of self-transformation through which they may learn to be modern subjects, with a capacity for moral, cultural, and political judgement) but would also, while claiming to extend the rights and entitlement of young people, actually diminish their unity-in-difference, their plurality, their particularity, and their non-identity with respect to the instrumental reason which drives the discourses of Youth Policies, from the nation-state (Nigeria) to the United Nations. For in the documents and discourses of these institutions, young people are hopelessly bureaucratized, rationalized, and turned into the totally administered objects of modernization and development. This is why the National Youth Policy and others of its kind are no more than the logic of substitution in which actually existing men and women are displaced by institutional discourse and reifying philosophies of Instrumentalism and Bureaucracy. Conclusion8. What is left of the concept of youth once it has been shown to be without reference to real subjects in the empirical world? The answer is: not a great deal. However, it is true that young people exist, that is, as human organisms, as persons, to whom intentional predicates can be ascribed. It is also true that young people, or youth (as a collective noun), have occupied specific yet contradictory positions in the discursive spaces of institutions from the family, the state, and the market economy to the “disciplines” of biology, physiology, sociology, psychology, and advertising. It is these “discourses” (as heterogeneous statements) that have produced and modulated our present image and conception of youth. These discourses have not only displaced the real, concrete persons that we may regard as young people, but have also produced the modern notion and meaning of youth. This is why, in the present discourse of youth, the assumed referent, young people or whatever youth may mean in reality, now no longer exists. We cannot now match the concept “youth” with the object; cannot, in other words, Ibrahim Bello-Kano 63 make the concept present to its referent. Thus it is the discourse of youth that is now the subject rather than youth itself (as a collection of concrete, living persons or individuals) which is, and has become, the object of the concept. In this sense, then, the word “youth” is no more than a signifier: it defers the meaning (of youth), and, at the same time, takes its place, thereby representing the absence (young people, the youth) that it evokes. As such, youth (the totality of the real persons) is now a representation (of discourse and of the signifier). To this extent, therefore, youth is, as an analytical or explanatory category, a vacuous concept, and without any analytic significance, for without it having some explanatory force, the concept dissolves into insignificance. And from the point of view of whether young people, youth, can be active agents in political, social, or national transformation, the best the discourse of youth can sustain is the politics of collective singularity in which the needs and interests of a subject represent those of others. But once we stress that one cannot assume young people as a single empirical subject, or as a normative subject that can be represented by one particular group or collective, then the whole conceptual and praxiological edifice of the discourse of youth collapses into a sign of incoherence. In the light of the foregoing arguments, youth cannot be an agency of, or for, neither can it (as a collective empirical subject) be mobilized for, an enduring political action— from changing the self-identity of human organizations to the changing of objective structures either for good or for ill. For if we were to ask the question, “in whose name should young people act?”, the answer, surely, would not consist in the proposition that it would be in the name of young people themselves, or youth itself — fundamentally because, from a logical and practical point of view, both the question and its possible answer are, as propositions, incoherent, meaningless. 64 Legon Journal of the HUMANITIES Volume 20 (2009) NOTES See the 1. National Youth Policy of Nigeria (2001). Abuja: Federal Government of Nigeria. I have also relied on the “Second Draft Review of the National Youth Policy of the Federal Republic of Nigeria”. (2008; unpublished).These two documents are substantially similar. This kind of critique is also called the “the triple critique” in that it unites conceptual 2. criticism and social-historical criticism, evaluation and explanation. It criticizes its object of inquiry in the very process of explaining it — so that the task of the critic is not just to interpret the object (false and dominatory beliefs and practices) but to change it (that is, to remove its conditions of possibility). Here the conceptual fuses into the historical, and vice versa: conceptual or historical change implies social and institutional change. As a method, meta-critique is an open, non-deterministic, analytic, for it offers the greatest range of real (non-utopian) possibilities that might be available to the critic. For more, see Bhaskar (1986) and Bhaskar (1993). It should be pointed out that physiologists do not use the concept of “youth” as such 3. but that of “puberty” to describe physiological changes in a “developing” human organism; while psychologists are likely to use the concept of “adolescence” to describe the mental-psychological process of becoming in an adolescent subject. This should be distinguished from the post-Romantic, post-Enlightenment philosophic 4. variety of self-creation championed by Nietzsche in which he suggests that we humans should become what we are in the sense of being able to invent ourselves. Here self- creation is to make a work of art of oneself. It is a “passion for self-reference” and self- fashioning, the will to give form to oneself, to be the author of one’s life. Nietzsche suggests further that “we have our highest dignity in our significance as works of art” (1967a: 52); that “truth is ugly… We possess art lest we perish of the truth” (1975: 435). According to Nietzsche, human beings want to be the poets of their life. For more on this, see Eagleton (2007) and Nehemas (1985). This principle is also suggested by the Principle of Charity; and both assume the 5. notion of a common human nature: that, in interpreting the beliefs and interest of subjects, we assume a tacit knowledge of the world and human subjects in general, for only then can we assume their rationality and the intelligibility of their doings and sayings, that is, as agents who are attitudinally and behaviourally rational. For more on this, see Callinicos (2007); Macdonald & Pettit (1981); Wiggins (1980). Ibrahim Bello-Kano 65 Davidson argues that since beliefs and meaning conspire to account for utterances, we 6. “cannot infer the belief within having the meaning, and have no chance of inferring the meaning without the belief” (1984: 142). According to Ernst Bloch (cited in Benhabib 1986: 353), social utopia is concerned 7. with human happiness, irrespective of any kind of difference between, within, or among human individuals, so that there is no place for the weary and the downtrodden. So by analogy, the utopia envisaged for youth is a negative one since it would not be a utopia at all but a circumscribed relativism based on age-identification, which would be incoherent in the discourse of social utopia, or even of natural rights, which assumes human dignity. In fact, according to Derrida (1978: 230), the 8. trace is “the erasure of selfhood” and of presence, one’s own presence, and is constituted by the disappearance of its disappearance. It is not, in other words, “a mortal germ” or a seed, or a substance— all of which are, by comparison, immobile and un-corruptible. In the discourse of youth, the real, young people, are never 9. sufficient in themselves, hence the reference, in these discourses, to extra-youth referents such as “development”, “national reconstruction”, etc. In this case, the very youth or young people, for whom the discourse speaks, become what Derrida (1976: 159) calls “supplements”, “substitute significations”. The real “thing” (youth) has vanished and now takes on meaning only as a trace (the discursive reconstructions of the discourse of youth) and in the invocation of the supplement (young people). In a word, youth has become, as an empirical subject, no more than a supplementarity (an addition-substitution) to an alien purpose outside the needs of young people qua young people. I have drastically summarized, and perhaps over-simplified, Foucault’s otherwise 10. elegant and sophisticated arguments. For a fuller treatment of the subject, and a critical review of the notions and concepts of discourse and power-knowledge, see Macdonnell (1986) and Mills (1997). This issue turns on the processes by which humans are constituted into subjects. 11. For Foucault it is through discursive practices. For Althusser (1971), individuals are formed into subjects within “ideology”. Yet I think neither proposition denies the causally efficacious reality of human beings, their being, by nature, what Bhaskar (1994: 105) calls “geo-historical products”. 66 Legon Journal of the HUMANITIES Volume 20 (2009) This has two senses: i) the notion of a subject, group, or human structure, as being the 12. foundation of knowledge, meaning, or an inherent purpose given by nature or natural determination; and ii) the notion of a particular group, class, or a collective, or even a particular man or woman, for example, Youth, Women, Men, the Proletariat, Party, Avant-garde Intellectuals, Third World People or Revolutionaries, or the Leader, as the universal hope of human emancipation, as particularity that represents universality. Ibrahim Bello-Kano 67 REFERENCES Althusser, L. 1971. Lenin and Philosophy and Other Essays. London: New Left. Belsey, C. 2002 Post-structuralism. Oxford: Oxford University Press. ________. 2005. Culture and the Real: Theorizing Cultural Criticism. London: Routledge. Benhabib, S. 1986. Critique, Norm, and Utopia: The Foundations of Critical Theory. New York: Columbia University Press. Bhaskar, R. 1979. The Possibility of Naturalism: A Philosophical Critique of the Contemporary Human Sciences. Brighton: Harvester Press. _________. 1986. Scientific Realism and Human Emancipation. London: Verso. _________. 1993. Dialectic: The Pulse of Freedom. London: Verso. _________. 1994. Plato, Etc.: The Problems of Philosophy and Their Resolution. London: Verso. Billington, R. et al. 1991. Culture and Society. London: Macmillan. Callinicos. A. 1989. Making History: Agency, Structure and Change in Social Theory. Oxford: Polity. __________. 2007. Social Theory: Historical Introduction. London: Verso. Davidson, D. 1984. Inquiries into Truth and Interpretation. Oxford: Clarendon Press. Dennett, D. 1981. Brainstorms. Montgomery, Vt.: Bradford Books. Derrida, J. 1976. Of Grammatology. Trans. G.C. Spivak. Baltimore: Johns Hopkins University Press. _______. 1978. Writing and Difference. Trans. Alan Bass. Chicago: Chicago University Press. Eagleton, T. 2007. The Meaning of Life. Oxford: Oxford University Press. Farrell, F. B. 1996. Subjectivity, Realism, and Postmodernism. Cambridge: Cambridge University Press. Fass, P.S. 1977. The Damned and the Beautiful. New York: Oxford University Press. Foucault, M. 1972. The Archeology of Knowledge. Trans. A. M. Sheridan. London: Routledge. _________. 1980. Power/Knowledge. Trans. C. Gordon et al. Brighton: The Harvester Press. _________. 1981. “The Order of Discourse”. Young, R. (ed.). Untying the Text: A Post-structuralist Reader. London: RKP, pp. 48-78. Fritz, S. 2005. “Youth”. Bennett, T. Grossberg, L. & Morris, M. (eds.). New Keywords. Oxford: Blackwell, pp. 380-82. ______. 1985. The Sociology of Youth. New York: Causeway Books. Giddens, A. 1981. The Class Structure of the Advanced Societies. London: Routledge. 68 Legon Journal of the HUMANITIES Volume 20 (2009) Gilles, J. (1974). Youth and History. New York: Academy Press. Gramsci, A 1971. Selections from the Prison Notebooks. Ed. Q. Hoare & G. N. Smith. London: Lawrence and Wishart. Hall, S. and Jefferson, T. 1976. Resistance Through Rituals: Youth Subcultures in Post-War Britain. London: Hutchinson. Hebdige, D. 1979. Subculture: The Meaning of Style. London: Methuen. Macdonald, G. & Pettit, P. 1981. Semantics and Social Science. London: Hutchinson. Macdonnell, D. 1986. Theories of Discourse. Oxford: Blackwell. Mann, M. 1986. The Sources of Social Power. Vol. I. Cambridge: Cambridge University Press. Mills, S. 1997. Discourse. London: Routledge. Mungham, G. and Pearson, G. 1976. Working Class Youth Culture. London: RKP. Nehemas, A. 1985. Nietzsche: Life as Literature. Cambridge, Mass.: Harvard University Press. Nietzsche, F. 1967a. The Birth of Tragedy. Trans. W. Kaufmann, New York: Vintage Press. __________. 1967b. On the Genealogy of Morals and Ecce Homo. Trans. W. Kaufmann, New York: Vintage Books. __________. 1975. The Will to Power. Trans. W. Kaufmann, New York: Vintage Press Saussure, F. 1974. Course in General Linguistics. London: Fontana/Collins. Wiggins, D. 1980. Sameness and Substance. Oxford: Oxford University Press. 69 GLOBALIZATION OF AN AFRICAN LANGUAGE: TRUTH OR FICTION? Josephine Dzahene-Quarshie1 Abstract In the quest to do away with every influence of colonialism including language imperialism, following the independence of several African countries, particularly in Sub-Saharan Africa, debates arose about the possibility of the adoption of Swahili as a common language for Africa since it was the most widely-spoken African language at the time. However, even though this discourse continued for some time, it did not go beyond this stage. In contemporary times, discourse on the Swahili language has emerged again, but this time it centres on the question of globalization of the Swahili language. This recent development therefore is the motivation for this paper. We seek to bring to light paradoxes that emanate from the debate and quest for the globalization of the Swahili language against the background of challenges that face the language at the national, regional and continental levels and to make submissions as to whether the quest for the globalization of the Swahili language is not just an illusion. 1. Introduction Following the independence of several African countries, particularly in Sub- Saharan Africa from the mid-nineteen fifties, debates arose about the possibility of the adoption of Swahili as a common language for Africa since it was the most widely-spoken African language at the time. The most notable propagators of this idea were the late Dr. Kwame Nkrumah, the first President of the Republic of Ghana, Professor Wole Soyinka, the Nobel Prize award-wining African writer, Professor Ayi Kwei Armah, author of the popular novel of the African Writers Series The Beautiful ones are not yet Born and Professor Ali Mazrui, a renowned Pan-Africanist, (Chimerah 2000, Mulokozi 2002). Though this discourse continued for some time, it did not go beyond the debate stage. Interestingly, in contemporary times, discourse on the Swahili language has emerged again, but this time it centres on the question of globalization and the Swahili language. This emerging discourse on globalization and the Swahili language is the main motivation for this paper. 1 Dr. Josephine Dzahene-Quarshie is a Lecturer in the Department of Modern Languages (Swahili Section), University of Ghana 70 Legon Journal of the HUMANITIES Volume 20 (2009) The second section of the paper sets out the background and justification for the paper. In the next section we give an overview of the current state of the Swahili language globally, that is at the national, regional, continental and global levels. We then examine the challenges that face the language at each level and their implications and highlight the emerging paradoxes for the question of globalisation. Finally, we re-examine the question of globalization and the Swahili language and draw our conclusion. 2. Background and Justification The term globalization has been defined as a ‘new term that is used to describe an old process, the historical process that began with our human ancestors moving out of Africa to spread all over the globe’ (Yale Global Online). This new term stems from the IMF, World Bank and World Trade Organization’s effort at achieving a global economy where indeed, there is free economic, political and environmental intervention. Tabb (2006) defines globalization as a ‘comprehensive term for the emergence of a global society in which economic, political, environmental, and cultural events in one part of the world quickly come to have significance for people in other parts of the world.’ He goes on to explain that ‘Globalization is the result of advances in communication, transportation and information technologies.’ Advances in these three areas have indeed turned the world into a global village. The term globalization for sometime has been used mainly in the economic sense, and has become a popular subject of debate and discussion in the business world and among scholars of economics but the term also encompasses other aspects of life such as culture and language. Indeed in whatever sense that the term globalization is used, communication plays a vital role. In other words, globalization cannot be achieved without communication. It is also a fact that communication is effected through the medium of language in whatever forms it may take be it vocal, written or electronic. In effect, there cannot be globalization without language. Some languages of the world already have been described or referred to as global or globalized. Typical of these is English. English is classified as such for various reasons, the most notable of which is the extent to which it is used globally. English is used globally in education, political administration and business. It is the major language that is used globally in publications including books, newspapers and magazines, on radio and television, in the movie and music industries. Although the Swahili language obviously cannot be equated with English there is no denial of its global presence. It would not be too far fetched to consider its Josephine Dzahene-Quarshie 71 possible globalisation along at least a couple of the general lines along which English is, for example, in the area of education and radio broadcast. Therefore, it is not surprising that Swahili has begun to receive some attention in the discourse on globalisation. In the literature two different views seem to have emerged as far as the globalisation of Swahili is concerned. On the one hand Swahili is seen basically as a language that possesses some qualities or characteristics that give it a high potential for globalisation (Moshi 2006, Mulokozi 2002), on the other hand Swahili is seen as having achieved globalisation to a limited extent (Kihore 2005). Moshi (2006) advocates the consideration of Swahili as one of the globalised world languages. She outlines the qualities that Swahili possesses that constitute potentials for globalization as the international recognition and status it has acquired over the years. She argues that the label ‘globalized’ should not be based on the size of population a language serves but rather on the functions it performs and the recipients of those functions. Based on her own definition of what a globalised language is she classifies Swahili as globalized on the grounds that Swahili’s presence is felt globally in the media (radio), ICT and academia. Mulokozi (2002) discusses the possible globalisation of the Swahili language against the background of its achievements as an international language. He suggests positive steps that ought to be taken in order to achieve this. Kihore (2005) on the other hand centres his argument on the globalisation of the teaching of Swahili. He explains that, so far efforts towards the globalisation of Swahili have been spearheaded by the main implementers of the process of globalisation (i.e. the developed world). He points out the importance of Swahili speaking countries taking up the challenge of ensuring the globalisation of the teaching of Swahili through their own efforts. Kihore like Mulokozi sees Swahili as having achieved globalisation as far as the teaching of Swahili is concerned, but his misgivings about the issue is that Swahili has been able to achieve this status mainly through the efforts of the super powers rather than the Swahili speaking countries. He argues that granted that globalisation has been described in some quarters as another form of colonisation, it is imperative that the needed steps are taken to ensure the reversal of roles and that the Swahili people themselves take over the task of ensuring further globalisation of the Swahili language by addressing and resolving the challenges that face Swahili speaking countries (and therefore hinders their efforts) such as inadequate ICT facilities, non existence of governmental and legal policies to back efforts of Swahili promoting institutions and the misconceptions and prejudices of the Swahili people that are detrimental to the efforts to achieve globalisation of the language. 72 Legon Journal of the HUMANITIES Volume 20 (2009) In the light of the above views on the globalization of Swahili we re-examine the question of globalisation of Swahili against the background of the challenges that face it and their emerging paradoxes. 3. An Overview of the Current State of the Swahili Language in Africa and Beyond 3.1 Tanzania In Tanzania the Swahili language has been the most important language for several decades. According to the 2002 population and housing census the population of Tanzania was about 34.5 million. The UN estimate of 2005 was 38.4 million. An estimated 90-96% of Tanzania’s population are said to speak the Swahili language proficiently either as a first or second Language (Mulokozi 2002, Rubagumya 1991). In fact for a greater percentage of the people, Swahili is their second language. However, there is a growing population of people who use Swahili as a first language most often as a result of their being brought up in urban centres. Swahili is used as the language of communication in all social, cultural, economic, political and administrative discourses on a daily basis and this has been so for many decades. Swahili was declared the national and official language, shortly after Independence. As a national language it is accepted by all and sundry and serves as a unifying force among Tanzanians. Kiswahili’s role as the official language of Tanzania is manifest in the socio-economic, political, and administrative settings. Over 20 newspapers are published in Swahili including those on sports and entertainment while only a few are published in English. There are several radio and television stations and again most of them broadcast in Swahili. Regarding the film industry, for a long time the local film industry was non-existent in Tanzania and this led to the influx of Western and later Nigerian films. The reason for this was that under President Nyerere’s rule, Tanzania did not have a national television company. However, presently the film industry is fast growing, and becoming popular. Swahili is the language used in these films, although it is often times code-mixed with English. So far Tanzania is the only country in Africa that has Swahili as both National and Official language and it has remained so for decades since the 1960s. As far back as 1962 that is shortly after attaining independence Swahili was also established by the ruling government (Tanganyika National Union) as the medium of instruction at the primary school level. Kiswahili remains the medium of instruction in all public primary schools. At the secondary school level, medium of instruction changes to English and Swahili becomes a compulsory subject, and Josephine Dzahene-Quarshie 73 at the tertiary level it is offered as an optional subject. Over the years several agencies and commissions have been set up by the government to see to the promotion of the Swahili language. Of these the most active and productive has been the Institute of Kiswahili Research (IKR), a research institute based in the Kiswahili Department of the University of Dar es Salaam. It was established in 1970. The institute which is made up of various sections, such as the literature section, lexicography section and linguistics section, has made an unqualified contribution towards the development of the Swahili language, amongst its achievements are, the periodic publication of list of vocabulary covering various areas spanning literary terms to technological and scientific terms and the publication of text books in Swahili in various disciplines and for all academic levels (Kiswahili 2005) 3.2 East Africa Next to Tanzania, Kenya has the largest population of native Swahili speakers in East Africa. In Kenya, Swahili is the first language of the natives of Mombasa a coastal town. It is also a lingua franca along with English in the urban centres. In the city of Nairobi both Swahili and English are used in most public and private institutions. In religious settings where the medium of communication is English most often interpretation into Swahili is offered. In Kenya, the official language is English and the national language is Kiswahili. In education, Swahili is a compulsory subject at the primary and secondary levels and optional at the tertiary levels. At the tertiary level Swahili is offered at both undergraduate and graduate levels. The national radio and television company, Kenya Broadcasting Corporation in addition to English, has Swahili slots. One newspaper Taifa Leo (The Nation Today) is the only state- owned newspaper published in Swahili. Kiswahili is quite wide spread in the Democratic Republic (DR) of Congo; it is spoken as a first language in provinces in North and South of the country and the eastern part of Upper Zaire by an estimated population of 9.1 million which is 27.49% of the population (Mputubwele 2003: 274). It is one of the four languages declared as National Languages in DR Congo. Radio Kinshasa and Voice of Congo broadcast some of their programmes in Swahili. In Uganda, Swahili has been and continues to be one of the Lingua Franca in its urban centres and in the military. Presently, Swahili’s presence is also felt in Rwanda and Burundi, it is spoken by a growing population. The state broadcasting companies have television and radio slots for Swahili medium programmes. 74 Legon Journal of the HUMANITIES Volume 20 (2009) Indeed, the East Africa community has adopted Swahili as one of its working languages, and in all the other countries bordering Tanzania, that is Sudan, Zambia, Mozambique and the Comoro Island there is a limited use of Swahili. 3.3 The Rest of Africa In the rest of Africa, Kiswahili is unknown to a lot of people. A yearly survey I undertake in my level 100 (first year) class indicates that less than 5% of the students know about the Swahili language and even fewer of them are able to tell where it is spoken. This is indicative of the extent to which people outside East Africa are oblivious of the Swahili language. Outside East Africa that is in the rest of Africa, Swahili at best would be found to be a subject offered at the tertiary level in a few Universities. Here in Ghana Swahili is taught in only two institutions, the University of Ghana and the SOS International School at Tema in Accra. In Nigeria Swahili is offered by the University of Ibadan and the University of Port Harcourt. Also the Voice of Nigeria has a Swahili service which has daily slots. Swahili is also studied in Egypt and Libya. It is evident that outside East Africa the presence of Swahili is not very visible; at best Kiswahili can be estimated to be studied by a few thousand students in the rest of Africa. The decision to adopt Swahili as one of the working languages of the Organisation of African Unity (OAU) (now African Union, AU) was taken in 1986 (CM/1352 (XLIV) but it was not implemented. The first four working languages of the OAU were English, French, Arabic and Portuguese. Although within the AU Swahili has been maintained as the fifth working language it has not been implemented. 3.4 State of Swahili in the Diaspora Interestingly, Swahili has a much stronger presence in the developed world especially in Europe and America. In these parts of the world the presence of Swahili is not felt in the number of speakers per say but in several other ways. First of all the teaching and study of Swahili is very popular. Because the American population is linguistically diverse, over the years as part of encouraging the study of minority and foreign languages, the federal government gives financial support for the teaching of such languages. Swahili has had an attraction for especially students of African descent who tend to identify with it. Several other African languages are taught in America but Swahili seems to be the most popular with students. It is taught in over 100 tertiary institutions. Swahili is also taught at the high school level. According to Online Resources (Digests 1996) 24 high schools in 14 states are involved in the teaching of Swahili. Josephine Dzahene-Quarshie 75 Almost all the universities that offer Swahili have websites where a lot of learning resources are made available with varying levels of accessibility. In Europe it is taught in universities in Britain, Germany, and Austria. In the Scandinavia Swahili is taught in Norway, Sweden, Switzerland, and Finland. In Asia it is taught in India, China and Japan. In the above countries a lot of researches on Swahili have been undertaken since the colonial period and many more are underway. Most of the world’s international broadcasting companies have a Swahili service. BBC, VOA, Radio Moscow etc., and there is abundant information on Swahili on virtually every website on the internet. 4. Challenges Facing the Swahili Language at Various Levels 4.1 National Level As much as Swahili scholars especially Tanzanian scholars glory in the status of Swahili as a potential global language, the current state of affairs where the Swahili language is concerned at the national level raises a serious question about the future of the Swahili language especially at the national, continental and ultimately the global level. In this section, our task is to bring to light these challenges that in our opinion stand to pose serious threats to the quest for the globalization of the Swahili language. A few decades ago, in the years following the independence of Tanzania the average Tanzanian took pride in the Swahili language and did not pay any particular attention to his or her inability to communicate effectively in the English language, but today the average Tanzanian is very keen about the acquisition of the English language. Awareness of the English language as a superior language has dawned on Tanzanians today. One major area of challenge stems from the mushrooming of ‘English medium schools’ all over the country. Shortly after independence, the ruling government gave a directive that Swahili should be the medium of instruction throughout the eight years of primary education, (Whitely 1969, Rwezaura 1993, Kiango 2005) and in reality this is so in all public schools. For several years after independence that is up to 1969 there was only one English/international school in Tanzania, established in Dar-es-Salaam, purposely to cater for children of expatriates, and Tanzanians who had lived outside the country for a long time. This was the only school in which the medium of instruction was English for obvious reasons. However by December 2005 the number of English Medium schools had shot up to 334.1 Most of these schools are non-government schools. 76 Legon Journal of the HUMANITIES Volume 20 (2009) The following graph shows the breakdown of English Medium private primary schools established during particular periods as against Swahili Medium schools.2 Table 1: Non-Government primary schools: English Medium versus Swahili Medium Of the 334 private, non-government primary schools only 19 of them have Swahili as a medium of instruction. From 1990 to 1999, 150 private schools were established and only 4 of them used Swahili as a medium of instruction. From January 2000 to December 2005, 186 schools were established and only 14 out of them use Swahili as the medium of instruction. This represents much less than 10% of the schools established within the period. These English Medium schools are distributed all over the country with higher concentrations in urban centres such as Dar-es-Salaam Arusha. The data indicates that most of the English Medium schools (over 95%) were established from 1990 to 2005. There is every indication that many more English Medium schools will be established and at a much faster rate and the possibility that they may pose a threat to Swahili medium schools cannot be ruled out. Moreover, English Medium pre-school day care centres are also springing up in the country. The picture painted above poses a great threat to the Swahili language although English medium schools represent a very small percentage of schools nationwide. 0 20 40 60 80 100 120 140 160 180 200 1960- 1969 1970- 1979 1980- 1989 1990- 1999 2000- 2005 NO. OF NON-GOV EMS NO. OF NON-GOV SMS Josephine Dzahene-Quarshie 77 Shortly after independence as part of the educational policy plan, Swahili was to begin as medium of instruction in primary schools while efforts were to be put in the preparation of educational materials in Swahili towards a later switch from English as medium of instruction at the secondary level to Swahili and eventually to the tertiary level. However, no concrete steps were taken by governments to see to the implementation of the ultimate plan of education. During the mid 1980’s there was a push from individuals to get the government to implement this plan, as the organs and agencies that were commissioned by the government to see to the preparation of materials and the working out of the change-over plan had completed the task. The government has been hesitant in implementing the use of Swahili at the secondary level as medium of instruction. However, it has not been bold to declare abandonment of this plan. Without the support of government, little can be done to defend the position of Swahili in Tanzania. Kiango (2005) states that: ‘In 1982, the presidential Commission on Education recommended a gradual shift of the medium of instruction in secondary schools from English to Swahili to begin from 1985 and to be completed by the year 1992.’ To date this is yet to be implemented and somehow, the impression that one gets is that Tanzanians’ view of the Swahili language has changed over the last few decades, now value is placed on the English language; they see the acquisition of English as very crucial to their future, and to their participating in the global world and economy. The apparent reason for the dramatic rise in the number of private English Medium schools and government’s reluctance to implement language policy plan concerning Swahili will be discussed later in the paper. The Institute of Kiswahili Research (IKR), based in the University of Dare- es-Salaam is one of the few institutions charged with the promotion of standard Swahili, and although it has achieved much in terms of publication of a wide range of materials including, dictionaries, vocabulary lists, literature books, grammar books, Swahili text books for all levels, even within Tanzania most people are not aware of its existence. A survey conducted by a member of the IKR itself indicated this (Nchimbi 2005). In the survey people of various professions and students were asked about their knowledge about the importance and contributions of the IKR to the growth and development of the Swahili language. The responses indicated that most of the ordinary people in Tanzania have no idea at all about the IKR (Ibid). The other Swahili promoting agencies have been the Directorate of Culture 78 Legon Journal of the HUMANITIES Volume 20 (2009) and National Language and the National Swahili Council which was established in 1969. The current trend in Tanzania where there is a desired shift to English medium at the primary level has the potential of shaking the very foundation upon which the future of Swahili in the country rests. The country may be heading for a major crisis where the future of Swahili is concerned. The hope of seeing Swahili used as the medium of instruction at the secondary and tertiary level and in the high court remains in question. Rubagumya’s (1991:68) statement that ‘the success of Swahili is very often exaggerated’ remains valid today. It seems there is a deliberate decision not to acknowledge the problem. 4.2 Regional Level Although Swahili is used in varying degrees in the countries bordering Tanzania, at the regional level (i.e. East Africa) there are no policies concerning the use of Swahili language. This is because each country in the Region has a different language policy, obviously tailored to suit its particular linguistic needs. Also some of them do not have clear language policies. Although the language and education policies in the other East African countries such as, Kenya, Uganda, D.R. Congo, Rwanda and Burundi at one time or the other make mention of the Swahili language, more often than not effort has not been made to see to their implementation. Typical of this is the case of Uganda. In 1973 under the government of Idi Amin, Swahili was declared (in speech only) as the national language of Uganda but this was neither documented nor implemented. Again in 2002 the Minister for Education announced that the teaching of Swahili in public schools would begin in 2003; however, this had not been implemented as at 2005 (Mukama 2005). There is also a lack of commitment to the selling of Swahili to the rest of Africa. Efforts at seeing to the promotion and consolidation of Swahili beyond Tanzania have been very minimal in our opinion, contrary to Kiango’s view (2005). 4.3 Continental Level Beyond East Africa, the influence of the Swahili language in the rest of Africa is very minimal and the major challenge facing the Swahili language at this level is that, again there is no one organization that brings Swahili teaching institutions outside the East African region together or promotes the language. As mentioned above comparatively very few institutions in Africa teach Swahili. The greatest Josephine Dzahene-Quarshie 79 challenge then is to see Swahili promoted and its position consolidated in the whole of Africa. 5. The Emerging Paradoxes From the above exposition, it is evident that Swahili has made strides as a taught foreign language since it is taught at some level of the educational ladder worldwide. In fact, over the last decade or so more and more academic institutions, especially, institutions at the tertiary level have taken steps to include Swahili in their curricular. The paradox then is that while Swahili is increasingly receiving global recognition and acceptance, Tanzanians are turning away from the language. The average Tanzanian today places a premium on the English language and believes that it must be acquired at all costs. This turn away from Swahili is evident in the number of English Medium Schools mushrooming all over the country. Several reasons may account for this gradual disinterest in the Swahili language particularly as a medium of instruction. However, in our opinion globalization is the main cause. It is evident that in this era of globalization, the English language occupies a central position. The retreat from the policy of Ujamaa (Socialism) could also be a factor to the drive towards English and westernization. Under Ujamaa, Tanzania was sheltered from the western world, but after the decline of Ujamaa in 1985, the move towards a more liberal economy which had been forestalled for so long was very swift, with Tanzania eventually succumbing to the World Bank and IMF structural adjustment policies (Maxon 1994: 258) and in time Tanzania opened up to the global economy and with it its vices one of which is the sudden awareness of the power of English. With the opening up of the Tanzanian economy, many foreign investors flooded the country, and proficiency in the English language undoubtedly became a requirement for employment into certain positions. Paradoxically, although the switch from English to Swahili medium at the secondary school level as a long term educational plan on the Tanzanian educational calendar is long overdue according to the records, there is reluctance on the side of government and policy makers to see to the implementation of this plan. The Tanzania National website gives information about educational reforms but makes no mention of the plan to change the medium of instruction at secondary level to Swahili. There seems to be no apparent reason why the changeover has not been effected. What then is to account for this reluctance of government to endorse and implement the changeover to Swahili? There seems to be two diverging opinions where this issue is concerned; one group is of the opinion that the policy of switching to Swahili medium will be 80 Legon Journal of the HUMANITIES Volume 20 (2009) the best for Tanzanian students and for that matter Tanzania as a country. The other group believes that English must remain as the medium of instruction indefinitely. Apart from the apparent pressure on government to maintain the English medium, part of the reason is that some of the people who are supposed to implement this policy are the very ones whose children would be found in private English Medium Schools. Of course parents who have paid dearly for their children to attend English medium schools at the primary level would not want them to switch to Swahili medium after all that effort? Paradoxically, some apparent promoters of the Swahili language and the elite in society send their children to English medium pre-schools and primary schools. Some go as far as sending their children abroad to acquire English medium education. In short Swahili is a victim of globalization; globalization is one of the major vices that wars against it. At the regional level although it is true that Swahili is gaining more grounds as the regional lingua franca, this has been due to natural evolutions such as population growth. No major efforts have been made to ensure the consolidation of Kiswahili in the region. The hope has been expressed that when the relatively newly formed East African Community which intends to have a common market and currency becomes fully operational its resolution to adopt Swahili as one of its official languages in its parliament will boost the use of the Swahili language further in East Africa Kiango (2005). Although this may be true, it may also act as an incentive for the acquisition of English by Tanzanians so that they may be able to compete with their Kenyan and Uganda counterparts who would be relatively more proficient in English. In our opinion the OAU now AU, could have played a much stronger role in promoting the Kiswahili language. The report of the UNESCO funded international Conference on language policies in Africa held in 1997 in Zimbabwe states that at this conference, African governments agreed to reactivate previous commitments made through the 1986 Language Plan of Action for Africa to make Swahili one of the working languages of the OAU. It is a fact that although Swahili has been adopted as the 5th working language of the OAU, it was never implemented. Again the AU which was established in 2002, confirmed Swahili as its 5th working language, but again that this had not gone past the adoption stage was made evident at the 2004 AU summit where, the President of Mozambique is said to have caused a stir when unexpectedly he decided to address the summit in Swahili. The AU of course was totally unprepared for this. Paradoxically no East Josephine Dzahene-Quarshie 81 African leader has addressed the summit in Swahili subsequent to this incident. One would have thought that East African leaders would have taken a clue from this incident and insisted on using the Swahili language at subsequent summits, but nothing of the sort has happened. Finally, it is paradoxical that efforts at ensuring that Swahili an African language thrives and is globally accessible are being championed mainly by foreign institutions. 6. Re-examining the Question of Globalization and Swahili Having discussed the current state of Swahili, the challenges facing it and the emerging paradoxes we now return to the question of globalization. In the introduction we pointed out that two main arguments seem to have emerged in connection with the globalisation of Swahili; namely, that on the one hand it is said to possess potentials for globalisation (Moshi 2006, Mulokozi 2002) and on the other hand it is said to have achieved globalisation in the Academia. To what the extent then can the language be said to be globalised and what conclusions can be drawn thereby? As indicated already, as an academic discipline and foreign language Swahili is offered on a global scale. Although in Africa, that is continentally where the Swahili should have been more widespread, very few institutions offer Swahili of course besides East Africa. In America Swahili is estimated to be taught in over a hundred colleges and universities. In America, Swahili is also taught at the high school level. According to an online resource, Digests (1996), Swahili is taught in 24 high schools. In terms of ICT, the Swahili language has had a great impact globally. Today through the internet Swahili is readily accessible to all in every corner of the globe as long as internet is available. So much information on the Swahili language is accessible from the internet right from vocabulary lists to lessons in grammar. Most of the universities that offer Swahili in the developed world have websites that make available learning materials and research findings (Kihore 2005, Mulokozi 2002, Moshi 2006). It is even possible to conduct searches in Swahili at some major web sites such as Google. At bbcswahili.com, a web site of the BBC, world news is made available in Swahili both in print and audio. In 2004, Microsoft launched a Swahili version of Microsoft Window software in Nairobi Kenya. In terms of radio broadcast, all the major international radio stations do have slots for Swahili broadcast, varying in time allocation. This ensures that Swahili 82 Legon Journal of the HUMANITIES Volume 20 (2009) natives wherever they are can benefit from these Swahili programmes. Learners of the language also benefit greatly from these. Looking at achievements of the Swahili language in the above areas it can be said to have achieved globalization not only in the area of the teaching of Swahili but also in the area of ICT and global broadcast. Nonetheless, in terms of global usage of Swahili as first, second, or even third language, Swahili is very far from achieving globalization. Although thousands of people may be learning Swahili or may have learnt Swahili relatively very few of them may actually use it on a regular basis. It cannot be compared with other global languages like English or even Spanish. According to Mulokozi (2002) Swahili is estimated to be spoken by about a hundred million people. This figure may, currently, well exceed a hundred million. Simple population growth could easily account for this. As far as global dimensions are concerned Swahili is very much confined to Central and East Africa and their environs, and a hundred million is just a drop in a global population of over eight billion. 7. Conclusion Taking the definition of globalisation by Tabb (2006) and others into consideration, Swahili can be considered to be globalised to the extent that it has, in the past, and still has a more or less global significance. However, we believe that if the needed steps are not taken to resolve the above paradoxes and to ensure the growth of the Swahili language at the national, regional and continental levels, the quest for the globalization of the Swahili language in the true sense of the word would remain but an illusion; the language would be limited to the classroom, ICT and the international media that are out of Africa’s control anyway. The quest for globalization in a fuller sense of the word will not be achievable in so far as the position of Swahili in Tanzania which is the main seat of Swahili in East and Central Africa remains threatened by government’s reluctance to implement pro-Swahili policies and English Medium Schools are left to take over Swahili Medium Schools and Swahili remains confined to Eastern and Central Africa as a lingua franca and efforts are not made by all stake holders to ensure the consolidation of Swahili in the rest of Africa in terms of Second Language or even Third Language Acquisition. In the area of teaching for instance in addition to Europe and America, native Kiswahili scholars could also help to strengthen the teaching of Kiswahili in the rest of Africa. Josephine Dzahene-Quarshie 83 Swahili may seem to be achieving globalization in a restricted sense of the word, but at the same time, the very foundation of Swahili in Tanzania in particular and East Africa in general is being threatened by the very nature of globalization and therefore concrete measures need to be taken to ensure that Swahili achieves globalization in a much fuller sense of the word than it has presently. 84 Legon Journal of the HUMANITIES Volume 20 (2009) NOTES The source of informatio1. n on English medium schools is the Tanzania National Website. www.moe.tz/pdf/regips.pdf EMS stands for Engl2. ish Medium School and SMS for Swahili Medium School. Josephine Dzahene-Quarshie 85 REFERENCES Blommaert, Jan, 1997 State Ideology and Language; The Politics of Swahili in Tanzania Published as LICCA paper nr.3 Duisburg; Ger hard Mercator University. Cleghorn A.; Merritt M.; Abagi J. O. 1989 Language Policy and Science Instruction in Kenyan Primary Schools, Comparative Education Re- view, 33, No.1: 21-39 Jay P. 2001 Beyond Discipline? Globalization and the Future of English, PMLA, Vol.116, No. 1: 32-47 Kiango J. G. 2005 Tanzania’s Historical Contribution to the Recognition and Promotion of Swahili, Africa & Asia, No. 5: 157-166 Kihore Y. M. 2005 Utandawazi na Ufundishaji wa Kiswahili Nordic Journal of African Studies,.14 (3): 384-390 Kimi K. 1998 The Paradox of US Language Policy and Japanese Language Education in Hawaii, International Journal of Bilingual Education andBilingualism, Vol. 1, No.1: 47-64 Kuntz P. 1996 African Language at the K-12 Level, Online Resources: Digests, 5 pp. Legère, Karsten, ed. 2002 Web version of Final Report by Herbet Chimhundu, Conference on Language Policies in Africa, Harare, Zimbabwe, 17-21 March 1998 UNESCO Mazrui, A.; Mazrui, A. A. 1993 Dominant Languages in a Plural Society: English and Kiswahili in Post-Colonial East Africa, International Political Science Review, Vol. 14, No. 3: 275-292 86 Legon Journal of the HUMANITIES Volume 20 (2009) Mazrui, A.; Mazrui, A. A. 1993 Dominant Languages in a Plural Society: English and Kiswahili in Post-Colonial East Africa, International Political Science Review, Vol. 14, No. 3: 275-292 Moshi L. 2006 The Globalized World Languages: The Case of Kiswahili Selected Proceedings of the 36th Annual Conference on African Linguistics, ed.Olaoba F. Arasanyin and Michael A. Pemberton, pp. 166-175. MA: Cascadilla Proceedings Project. Mukama R. 2005 Mchangao wa Taasisi ya Uchunguzi wa Kiswahili (TUKI) Katika Uendelezaji wa Kiswahili Kiswahili Vol. 68: 103-114 Mulokozi M. M. 2002 Kiswahili as a National and International Language Web publication on Http://www.helsinki.fi/hum/aakk/documents/ kiswahili.pdf (Asian and African Department, University of Helsinki. pp1-11 Omar A.S.; Sanneh S. 2002 African-Language Study in the 21st Century: Ex pansion through Collaboration and Technology, African Issues, Vol. 30, No. 2: 42-49 Ornstein J. 1964 Africa Seeks a Common Language The Review of Politics, Vol. 26, No.2: 205-214 Rothkop D. 1997 In Praise of Cultural Imperialism? Effects of Globalization on Culture, Foreign Policy, 9 pp. Registered Non-Government Primary Schools as of Dec. 2005….15/4/2005. S.L.P 55139. English Medium. www.moe.tz/pdf/regips.pdf Stevick E. W. 1967 The Teaching of African Languages in the United States since 1961, African Studies Bulletin, Vol. 10, No. 1: 16-21 Tabb, William K. 2006 Globalization, Microsoft Encarta Online Encyclopedia 2006 http://encarta.msn.com/text_1741588397___1/ Globalization.html Josephine Dzahene-Quarshie 87 Tanzania National Website http://www.tanzania.go.tz/human.html The FW de Klerk foundation, 2007 The Challenges: Co-Existing in a Globalised World, 6 pp. Yale Global Online Yates B. A. 1980 The Origin of Language Policy in Zaire The Journal of Modern African Studies, Vol. 18, No.2: 257-279 89 CORRUPTION AND THE BODY POLITIC IN POST-COLONIAL GHANA: A RE-READING OF AMU DJOLETO’S MONEY GALORE IN THE ERA OF ‘ZERO TOLERANCE FOR CORRUPTION’ IN GHANA Mawuli Adjei1 The native intellectual who wishes to create an authentic work of art must realise that the truths of a nation are in the first place its realities. - Frantz Fanon, The Wretched of the Earth Abstract Corruption has been an important subject of analysis by social scientists for many years (Bayley 1966; Huntington 1968, 1990; Gould 1980, 1989; Ali 1985; Crowder 1987; Kimenyi 1987; Alam 1989; Mbaku 1991; Couch et. al. 1992 etc.). In the 1960s, however, two major events rekindled interest in the study of corruption, especially in developing countries. First, Samuel Huntington and others developed theories of modernisation and political development. Second, the economies and markets of the newly developed countries of Africa and Asia were overwhelmed by corruption, bureaucratic inefficiency and incompetence. Apart from social scientists, creative writers have also exposed and dramatised the massive spectre of corruption in Africa. Like other urban political novels, Amu Djoleto’s Money Galore (1975) vividly captures the manifestations, scope, function, psychology, power and cultural imperatives of corruption in post-independence Ghana. 1. Introduction Many studies have been conducted on corruption in Ghana (Apter 1963; Owusu 1970; Gyekye 1977; Levine 1975; Chazan 1983; Diamond 1987; Baynham 1988; Ayittey 1992). In all these studies Ghana was seen as plagued by high level unethical behaviour since the colonial period. This includes bribery, embezzlement, fraud, extortion and nepotism. As far back as 1948, the Watson Commission hinted that it would be idle to ignore the existence of bribery and corruption in many walks of life in the Gold Coast and warned that it might spread as further responsibility devolved upon the African in the context of self-rule and bureaucracy. 1 Dr. Mawuli Adjei is a Lecturer in the Department of English, University of Ghana. 90 Legon Journal of the HUMANITIES Volume 20 (2009) During the several military and civilian regimes in the history of Ghana, the citizens have come to terms with corruption and its consequences as indicated in the Final Report of the Commission of Enquiry on Bribery and Corruption in Ghana. The Commission actually identifies corruption as a crime (Ayittey 1992). Corruption was a major campaign issue in the 2000 general elections. The then ruling National Democratic Congress government, headed by Jerry Rawlings, was perceived as corrupt. The largest opposition party at the time, the New Patriotic Party, promised to run a clean government. In its manifesto, dubbed Agenda for Change, the party pledged, inter alia, to reduce “the potential for, and level of, corruption in the country through removing rules and regulations that create rigid bureaucracy and provide opportunities for corruption” It also promised to “restore credibility to the executive branch of government by setting a higher moral tone by example and precept” and ensure that “public servants execute their assignments with honesty, transparency, discipline, equity and accountability” (pp.37-38). Indeed, upon assuming office, President John Kufour (President of Ghana, 2000-2008) pledged “ZERO TOLERANCE FOR CORRUPTION.” In recent times, for three years running, Ghana has faired badly on Transparency International’s Corruption Perception Index. In 2004, Ghana, with a score of 3.3 (out of 10), ranked the 64th most corrupt nation on earth. In 2005, she ranked 65th with a score of 3.6. In 2006, she ranked 70th with a score of 3.3. As Linda Ofori-Kwafo, Programmes Manager of the Ghana Integrity Initiative, lamented, “As Ghanaians, we should be concerned with Ghana’s 2006 CPI score because it authenticates the findings of local surveys conducted in Ghana in 2005.” According to her, “The Centre for Democratic Development (CDD) Ghana’s 2005 Afro Barometer Round 3 findings indicated that popular perception of corruption among public officials is apparently on the rise.” And on April 23, 2007, The Daily Graphic quoted Professor Henrietta Mensah-Bonsu (a Professor of Law in the University of Ghana) as saying that “We ourselves participate in [corrupt] practices in order to secure advantages for ourselves or cause officials to short-circuit systems for our benefit.” She made this observation at the St Peter’s Regional Seminary at Koforidua on ‘Contemporary Societal Attitudes towards the Promotion of Justice and Reconciliation: Prospects and Challenges.’ It is significant that this observation was made among the clergy, men and women whom the populace look up to as the gatekeepers of public morality, but who are themselves embroiled in unethical behaviour. All these, in addition to the numerous Auditor-General’s Reports presented to the Public Accounts Committee of Parliament, show a strong indication that Ghana has developed what could be termed a culture of corruption. According to Chazan (1980:86), Mawuli Adjei 91 It had been long in the making, but then its outlines were unmistakable. Bribery, graft, nepotism, favouritism, and the like had become commonplace at all levels of officialdom; and what is more, much of the public had come to expect officials to conduct their business in a spirit of subterfuge, dishonesty, and mendacity on all sides. How are the perceptions of corruption captured in Amu Djoleto’s Money Galore? The writer sees corruption as grounded in Ghana’s politics and governance, cultural values, poverty and weak morals. In Money Galore, Ghana is seen as a ‘kleptocracy’; that is, a political system operated by thieves. To put the novel in its proper perspective one needs a brief insight into politics and governance in Africa in general and Ghana in particular. In Africa disrespect for law and procedure, impunity, arrogance, political party organization, incompetence, victimization, personality cult etc. combine to perpetuate an order of blatant corruption. The nature of politics in post-colonial Africa is such that the ruling elite promote a system of inequality through accumulation of power and wealth. As Bayart (1992) rightly puts it, In Africa…the State is a major manufacturer of inequality. The development which it boastfully claims to promote, and in whose name it attempts to ban political competition and social protest, plays a part in the process. Indeed, across Africa, popular representations introduce different nuances to the construction of inequality. Although in Ghana these nuances are not so well structured in public parlance, as Price (1974) has observed, the big man/small man syndrome defines and determines access to power and wealth. This is best demonstrated in the arena of politics. Those who enter politics do so because it is the easiest access to wealth and power. 2. Corruption in Money Galore In many urban political novels by African writers, including Chinua Achebe’s No Longer at Ease (1960) and A Man of the People (1966), Kofi Awoonor’s This Earth, My Brother… (1970), Ayi Kwei Armah’s The Beautyful Ones Are Not Yet Born (1964), Alex Agyei-Agyiri’s Unexpected Joy at Dawn (2004) and the subject of this article, Amu Djoleto’s Money Galore, people enter politics because of monetary and material gains. According to Cartey (1971:158), 92 Legon Journal of the HUMANITIES Volume 20 (2009) Many of the characters and types enter politics specifically for its lure, for the momentary glitter and glamour that accompany it and for the prestige and power it gives them. Entry into politics seems to destroy morality of any kind, transforming the politician into a self-seeking and unscrupulous activist. As we indicated above, it would be wrong to study Money Galore in isolation. It is a novel which adds to the compendium of post-colonial African novels which depict the post-colonies, whether in West Africa or East Africa, as effectively mired in corruption and mismanagement. The main character in Money Galore, Abraham Kafu, is a disgruntled and frustrated secondary school teacher. Getting on in years and with no immediate prospect of promotion and the benefits that go with it, he opts for politics. Thus, Kafu’s entry into politics is predicated mainly on his desire to make money and to make it big and fast. On the verge of entering politics, he tells his wife: We need a house pretty soon. It shouldn’t take more than six months. Then I’ll have a home for the first time! I’ll be a minister! I’ll be rich! Oh, gorgeous! (p.38). Kafu knows exactly what he is talking about. Being a minister comes with all the perks of office, power to do anything, including changing the rule of law with impunity, award of fraudulent contracts, access to all the beautiful women and the good life. To do this, he would have to liaise with crooked business men and women. Notable among these people are Nee Otu Lartey, Anson Berko, Odofo and Salamatu. These people are able to worm their way into every government, civilian or military. They represent the supply side of corruption. Take Lartey, for instance, who made his fortune as a foreman in the Ministries: As supervisor, he displayed tremendous resourcefulness. He was responsible for the maintenance of government quarters and bungalows and began to divert large quantities of materials into his private depot in Nima, Accra. All he had to do was to over-requisition stores, or, instead of replacement, he would order repairs and again divert the stores (p.21). Through this, he manages, after five years as supervisor, “to put up a three-storey block of flats to let and a comfortable six-room, one-storey house to live in himself” (p.24). Mawuli Adjei 93 Salamatu and Odofo are Makola (market) women who know how to keep their businesses going. They bargain especially with their bodies. Odofo, a textile dealer, “was operating passbooks with G.B Ollivant, UAC, GNTC, UTC, SCOA and CFAO” (p.46). These women can smell a new political client and ally from afar. On Kafu’s maiden visit to Makola, even as a mere candidate, Many of them recognised Kafu and rushed on him. Shouts of ‘our darling!’, ‘my sweetheart!’, ‘my husband!’… filled the air. They embraced him, some removed their cover-cloth to fan him like a chief, some sprayed him with lavender and a daring eighteen- year old girl poured white powder on his head and tied a white calico round his neck …(p.46). Djoleto expresses, through the conduct of the central character, Kafu, and his accomplices, his disillusionment with post-independence Ghana where corruption and political ineptitude and repression have become the defining features of the national ethos. He demonstrates that the emergent Ghanaian administrative and political elite who have inherited power from the colonial masters have instutionalised a system of pillage and people, high and low, have been socialised and acculturated to see the state as the ultimate trophy to be competed for individually or in small corporate cabals. In Money Galore, Djoleto presents a state in which central authority is hardly visible. Vuga and Mensah who represent the establishment and the state apparatus, and who initially come across as scrupulous exemplars of the Civil Service ideal, are, in practice, at heart, self-seekers and white-collar thieves. They have outlived every administration and have become conversant with the rottenness that pervades the entire fabric of government. Apart from these two characters, the entire state apparatus is embodied in Kafu, Minister of Social Welfare, and the world of the novel is refracted through the personality of this single character. He appears to be the one man who exercises power on behalf of the Executive, Legislative and Judicial arms of government. The picture one gets in Money Galore is one of a country close to a failed state, questioning, as it were, the capacity of Ghanaians to rule themselves effectively. Although the novel was published in 1975, its timelessness is seen in how the issues explored are germane to today’s Ghana. Today, Ghana may boast of a Parliament of 230 representatives and not a single Abraham Kafu who makes and unmakes laws; the country may have a Judiciary that purports to be independent and dispensing justice in accord with the rule of law; the country may have a 94 Legon Journal of the HUMANITIES Volume 20 (2009) visible president who is exercising his mandate on behalf of the people; the country may have designed and firmed up, somewhat, the structures and institutions of democratic governance, yet the harsh reality is that Ghana is still a very, very corrupt country. 3. Money Galore and Social Change Why is it that twenty-nine years after The Beautyful Ones and twenty-two years after Money Galore Ghana is still a corrupt country? Why is political corruption a major concern in Ghana today? To answer these questions, with specific reference to Money Galore, we need to first delve into the issue of social change. If we regard Money Galore as a novel of social change, to what extent does it provoke or promote social change? What is social change, to start with? Sociologists and anthropologists (Moore 1963; Dahrendorf 1977; Jeffreys 1992) have identified exogenous and endogenous conflicts in any society. Exogenous influences come from outside while endogenous influences come from within. But we must also distinguish between the readily observable and the not so readily observable effects of social change in a given society. According to Cancian (1978), change within the social system refers to change that does not alter the system’s basic structure. We must also note that social anthropologists have divided the culture of societies into two broad categories – material culture and non-material culture. The material culture embraces the structure of the society (government and institutions), its artefacts and its technical achievements, which can easily be identified. They constitute the visible manifestations of a society’s existence. Change in this area is considered change of the system because it alters things which can be readily observed, measured and catalogued. On the contrary, the non-material culture (primary) cannot be readily observed. It can neither be measured. This embraces the traditions, beliefs, ideas, worship, values and ways of life of a whole people. The question then is: Is Djoleto advocating exogenous or endogenous change, a change of material culture or non-material culture? The answer to this question lies in the writer’s technical approach to the issue of corruption in Ghana. That is, a writer’s interest in and attitude to critical issues in a novel of social change are seen in the special techniques he employs. This is underscored by Agovi (1988: 12): The techniques are special…because the phenomena of social change impinging on the imaginative novelist, owing to the emotional reactions which they arouse in him, give rise to techniques of peculiar complexity which, it is further suggested, Mawuli Adjei 95 increases in frequency and in intensity as the inherent ambiguities of social change are analytically exposed. Djoleto’s approach to the exposition and critique of corruption is satiric, symbolic, ritualistic and cyclical. As a satire, Money Galore highlights the rottenness in the Ghanaian body politic. The society literally stinks because corruption cuts across the whole spectrum of the body politic – politicians, the clergy, civil servants, men and women. Those who are visibly corrupt are caricatured and ridiculed because they are cast as unscrupulous and morally degenerate. The symbols employed by Djoleto amply and vividly reflect the social malaise. Many of the metaphors in the novel centre on rottenness, filth and putrefaction, metaphors which have been carefully chosen to provoke revulsion to corruption. As Wright (1989) notes, Filth and shit, which have a way of finding out fraud and guilt and bringing the powerful back to the squalor out of which they have corruptly carved a niche of cleanness and prosperity, function as levelling metaphors which put down pretension, expose false gentility and attack corrupt leadership. Kafu, the central figure, is MP for Accra Central. Accra is not only the capital of Ghana but also the centre of political corruption since it is the centre of government. Accra Central appears to be a well-chosen microcosm of what government and governance symbolise. Besides, Accra Central is where the famous Makola market is located. So symbolic and iconic was Makola as the hub of corruption in the 1970s that during Rawlings’ 1979 AFRC Revolution, the entire market was razed to the ground. Thus, in Money Galore, those who are the architects of corruption are connected directly to Makola and Accra Central. Take, for instance, Odofo and the environment in which she operates at Makola: Her stall was eight feet by twelve. It was right in the thick of the filth of the market. There was a small drain eight inches deep running in front of it. It had been cleaned the evening before but had now been filled up again with dead mice, rotting vegetables, fish and urine…Dirt reigned all over and the awful smell of dried fish permeated the whole area. The Gaos who sold the grains had not washed for a week. They lay idly on the bags of grains, picking their noses and scratching their arses (p.38). Another symbol which represents the rottenness in the society is the premier Teaching Hospital of the country, Korle-bu. The hospital, which is supposed to 96 Legon Journal of the HUMANITIES Volume 20 (2009) restore the citizenry to health, is, physically, as rotten as Makola. After surviving an accident Kafu finds himself at Korle-bu and cannot help complaining that This place is just like Makola…The place was narrow, dingy and crowded. It was a congestion of the rag-tag of humanity… frightening tuberculosis patients, cardiac patients whose breathing scares you, all sorts of cases, both strong and dying – you feel you’re trapped – all your ego goes (p.36). Djoleto also sees corruption as both ritualistic and cyclical in the sense in which the national psyche and the national institutions are perpetually conditioned and predisposed to corruption. So long as access to power is driven by money and the people have come to accept money as the sine qua non, moral values have no place in the collective psyche. Those who have not properly internalised this inverted value become endangered species and outcasts. A clear example is Adjin- Yeboah. According to Nee Otu, Adjin-Yeboah never properly understood money because “Where there are free elections, no one gets power by sheer moral appeal” (pp.56-57). So money is the ritual item that lubricates the so-called democratic process and it functions in perpetuity as long as people hanker after political power or public office. Another damning revelation comes from Anson Berko. In the heat of his conflict with Kafu, even though he is officially sacked as the bread contractor for Ghana National, nothing changes. Berko says: This school will have Kafu’s own man as the next bread contractor for the school out of the largeness of his heart. I am going, but another vulture (my emphasis) will soon land on this compound (p.70). In spite of situating the novel within the context of physical filth and moral degeneracy, and depicting corruption as ritualistic or cyclical, the novel is very weak in constructing a framework for social change. Change does come at the end of the novel but it is basically change of the material culture and change of the system. As change of the system, it merely concentrates on regime or institutional change and change of individuals. Such a change can only be cosmetic and, in essence, constitutes no change at all. It substitutes one corrupt system, and all its paraphernalia, for another corrupt system. This is a familiar motif in Achebe’s A Man of the People and Armah’s The Beautyful Ones where the military oust corrupt civilian regimes. In both novels the military amply demonstrate that they Mawuli Adjei 97 can be more corrupt than the regimes they have removed. In practice, those who take over come with the primal vampire culture of repression and looting of the national coffers. Money Galore buys directly into this cosmetic change. There is a military coup in the end. In fact, it is widely rumoured long before it occurs. The rumour of a coup, rather than induce fundamental change in attitude on the part of the corrupt, galvanizes them into panic pillage. Kafu, Berko, Lartey, Odofo and Salamatu move to grab as quickly as they can. As Otu Lartey unabashedly confesses, I am a businessman, and I spend long sleepless nights speculating. For any little investment, I must make the maximum profit before some lout like Kafu wrecks my entire enterprise – by a stroke of the pen (p.70). This is not lost on the principal character, Kafu, himself. When Odofo confronts him with the prospect of an imminent coup d’etat, he minces no words about his intentions. To Odofo’s question, “So, granted it’s going to happen, how are you prepared against it? What are you going to do? he replies: “Money…Yes, money. I must garner lots of it before the blow falls” (p. 166). For a society which is suffering from corruption of the body politic in terms of corruption of values, corruption of the mind, corruption of the body and corruption of the soul, there is no fundamental change in the psyche and perception of the suffering masses. All they want is the overthrow of the vampire state and those who have come to embody or symbolise it. The novel’s major weakness, therefore, is its total lack of a moral centre - a character or a group of people - committed to initiating fundamental change philosophically or ideologically; people who must be seen to be atypical and whose actions and thoughts would constitute a counterbalance to the status quo. We have seen the failure of individuals like The Man and Teacher in The Beautyful Ones, Obi Okonkwo in No Longer at Ease and Odili in A Man of the People who do make statements of intent but are frozen in inaction and rhetorical ambiguity. No character in Money Galore comes close to even these failed pseudo-revolutionaries. The main problem, it has to be conceded, is embedded in the country’s cultural values and poverty. The masses, it appears, are part of the problem. Due to inertia, moral degeneracy, promiscuity and double standards – what Armah terms the ‘chichidodo syndrome’ - those caught in the middle of the corruption nexus – neither on the supply nor demand side – are unable to fight corruption. Although sometimes they are not directly involved in corruption, they indirectly 98 Legon Journal of the HUMANITIES Volume 20 (2009) foster it through their inaction and hypocrisy. In most cases, they benefit directly from it. Benjy Baisi, headmaster of National Secondary School, pretends to be helpless when the bread contractor, Anson Berko, schemes to reduce the weight of the bread supplied to the school. Baisi protests vaguely. Then Berko throws a bait at him in the form of a loan on very soft terms to build a house for him. Baisi falls for the bait. A personal house, regardless of how it is acquired, is more important than the fraudulent reduction of the weight of the bread for the students. Yet, so many months after, the same Baisi complains to Berko in a rather sanctimonious manner, in an attempt to dissociate himself from the deal. He says: [B]ut the thorn in my conscience is that you shouldn’t ever have given me or my wife any money when you were reducing the weight of the bread. This was your crime and I think I was your partner (84). The truth of the matter is that Berko strikes a chord deep within the cultural mindset – the need to impress the public through an impressive funeral and to bequeath a legacy, especially landed property, to your children and the extended family, no matter how it is acquired. Berko tells Baisi, Benjy, use your head. You’re the headmaster of a big secondary school…You are old and you own no house. When you die there will be nowhere to lay your body in state unless you’re put in Anomabu Town Hall, having been carted in a deep freezer from Cape Coast (pp. 5-6). Thus, Baisi’s inability and lack of moral courage to resist fraud and corruption are driven by considerations that have strong cultural foundations. The society expects that the headmaster of a big secondary school must die big, lie in state in a big house and be buried in grand style. Similarly, one of the major characters in the novel, Amega, confides in Kafu: I have four children, two boys and two girls. My ambition is to leave a house to each of them some day. This ambition is within reach. I’ve started the fourth one and I need money to complete it (p.158). Mawuli Adjei 99 It would appear, then, that the extended family system, for instance, plays an important part in the ways in which people hanker after material possessions, especially money. Issues relating to inheritance, funerals and landed property drive people to acquire money by any means possible. These personal ambitions, grounded on an unbridled quest for money, constitute the real motive for corruption in the novel and why the novel is titled Money Galore. These ambitions stink. Towards the end of the novel, Djoleto chooses a very symbolic professional group to press home the point of collective moral degeneracy and the masses’ complicity in it. He uses the conservancy labourers and their leaders to demonstrate how money has become an idol for individuals and groups. Their profession stinks; so do their leaders. In the midst of an industrial action, when they meet Kafu to negotiate for better conditions of service, it takes very little effort for Kafu to bribe their leader and scuttle the strike action as the scenario below reveals: Kafu sent for their chief, who being alone with him was completely subdued and had become cringing. The setting of the office and the ministerial presence contributed greatly. Kafu spoke softly, in a conciliatory and very friendly fashion. ‘Now see, my friend, you scratch my back, I scratch your back, you understand?’ ‘Yes, sir.’ ‘What do you want?’ ‘Justice, sir. Even you add fifty pesewas to our daily rate, we wouldn’t mind, sir.’ ‘My friend, don’t talk about justice. Nobody eats justice. You want money, I’m sure?’ ‘Yes, sir.’ (p.146) The conservancy labourers know they are poor, yet they are not interested in finding out why they are poor. They have come to accept the fact that Kafu and the like are corrupt and rich. But their understanding of democracy is ‘One Man No Chop.’ All they want, in the context of the eternal big man-small man equation, is a small slice of the national cake. They are not aware of the linkage between corruption and poverty. Are all these issues raised in Money Galore relevant to Ghana’s poverty levels today? When Ghana declared itself a ‘Heavily Indebted Poor Country’ (HIPC), was there any correlation between the country’s HIPC status and decades of corruption? As Transparency International notes, 100 Legon Journal of the HUMANITIES Volume 20 (2009) A strong correlation between corruption and poverty is evident in the results of the CPI 2006. Almost three-quarters of the countries in the CPI score below five (including low-income countries and all but two African countries) indicating that most of the countries in the world face serious perceived levels of domestic corruption (TI Corruption Perception Index, 2006, p.2). It would appear that the structures of corruption are so well laid that the total fabric of society functions, consciously or unconsciously, through these structures. Transparency International diagnoses the system aptly, noting that: Corrupt intermediaries link givers and takers, creating an atmosphere of mutual trust and reciprocity…to provide legal appearance to corrupt transactions… (TI Corruption Perception Index, 2006, p.3). 4. Conclusion What is the way forward? We have indicated that there is not a clear path constructed in Money Galore. The Declaration of the 12th International Anti-Corruption Conference in Guatemala City, Guatemala, November 18, 2006, themed ‘Towards a Finer World: Why is Corruption Still Blocking the Way?’ emphasises respect for values and especially education, “because children must understand and respect core humanitarian values if the long-term fight against corruption is to be won” (TI, p.3). It is doubtful if this would amount to much in Ghana’s circumstances. Ghana’s corruption, it appears, is rooted in the body politic, and the people’s body language does not appear to be anti-corruption. The structures that foster corruption appear to be ingrained in the national psyche. Mawuli Adjei 101 REFERENCES Achebe, Chinua. 1966. A Man of the People. London: Heinemann. 1960. No Longer at Ease. London: Heinemann. Agenda for Change: Manifesto of the New Patriotic Party, 2000 Agovi, K. E. 1986. Novels of Social Change. Accra: Ghana Publishing Corporation. Apter, D.E. 1963. Ghana in Transition. New York: Atheneum. Alam. 1989. “Anatomy of Corruption: An Approach to the Political Economy of Underdevelopment.” American Journal of Economics and Sociology 48(4), pp.441-56. Ali, M.S. 1985. Corruption: A Third World Perspective. Lahore, Pakistan: Aziz Publishers. Armah, Ayi Kwei. 1964. The Beautyful Ones Are Not Yet Born. London: Heinemann. Awoonor, Kofi. 1970. This Earth, My Brother…London: Heinemann. Ayittey, G.B.N. 1992. Africa Betrayed. New York: St. Martin’s Press. Bayart, Jean-Francois. 1993. The State in Africa: The Politics of the Belly. London/ NewYork: Longman. Bayley, D.H. 1966. “The Effects of Corruption in a Developing Nation.” The Western Political Science Quarterly 19(4), pp. 719-32. Baynham, S. 1988. The Military and Politics in Nkrumah’s Ghana. Boulder: Westview Press. Cancian, Francesca. 1978. Functional Analysis of Social Change. Oxford: Blackwell Publishers Ltd. Cartey, Wilfred. 1969. Whispers from a Continent: The Literature of Contemporary Black Africa. London: Heinemann. Chazan, Naomi. 1983. An Anatomy of Ghanaian Politics: Managing Political Recession, 1969-1982. Boulder: Westview Press. Crowder, M. 1987. “Whose Dream Was It Anyway? Twenty-five Years of African Independence.” African Affairs 86(342), pp. 7-24. Cummings, Jeffreys. 1992. The Anthropological Perspectives. London: Heinemann. Dahrendorf, Ralf. 1977. Towards a Theory of Social Conflict. New York: Longman. 102 Legon Journal of the HUMANITIES Volume 20 (2009) Djoleto, Amu. 1975. Money Galore. London: Heinemann. Diamond, Larry. 1987. ‘Class Formation and the Swollen State.’ The Modern Journal of African Studies 25 (4), pp. 567-590. GII Alert (Ghana Integrity Initiative Quarterly Newsletter), No.12, December 2006. Gould, D.J. 1980. Bureaucratic Corruption and Underdevelopment in the Third World. New York: Program Press. & Mukendi, T.B. 1989. Bureaucratic Corruption in Africa: Causes, Consequences and Remedies. International Journal of Public Administration. Gyekye, Kwame. 1977. Political Corruption: A Pilosophical Inquiry into Moral Problems. Accra: Sankofa Publications Ltd. Huntington, S.P. 1968. Political Order in Changing Societies. New Haven: Yale University Press.1990. “Modernization and Corruption.” In Heidenheimer et al. (eds.) Political Corruption: A Handbook. New Brunswick, N.J: Transaction Publishers, pp. 375-88. Kimenyi, M.S. “Bureaucratic Rents and Political Institutions.” Journal of Public Finance and Public Choice 3, pp.39-49. Levine, V.T. 1975. Political Corruption: The Ghana Case. Stanford: Hanover Institution Press. 1993. ‘Administrative Corruption and Democratization in Africa: Aspects of the Theoretical Agenda.’ Corruption and Reform 7, (3), pp. 271-278. Mbaku, J.M. 1994. ‘Bureaucratic Corruption and Public Policy Reform in Africa.’ Journal of Social, Political and Economic Studies 19 (2), pp. 149-75. 1996. (Ed.) Corruption and the Crisis of Institutional Reform in Africa. Lewitson, UK: The Edwin Mellen Press. 1999. ‘Bureaucratic Corruption and Africa: The Futility of Cleanups.’ The Cato Journal (Spring). Available online at www.epnet.com/bin/epwgargoy. Mawuli Adjei 103 Moore, Wilbert. 1963. Social Change. Prentics-Hall Inc. Owusu, M. 1970. Uses and Abuses of Political Power: A Case Study of Continuity and Change in the Politics of Ghana. Chicago: The University of Chicago Press. Price, R. 1974. ‘Politics and Culture in Contemporary Ghana: the big man-small man syndrome’, Journal of African Studies (Summer), 1 (2), pp.173- 204.1973. ‘The Pattern of Ethnicity in Ghana: A Research Note.’ Journal of Modern African Studies, XI, 3, pp.470-475. Shillington, Kevin. 1992. Ghana and the Rawlings Factor. London: Macmillan ‘Transparency International Corruption Perceptions Index 2006’. Available online at press@transparency.org Wright, Derek. 1989. Ayi Kwei Armah’s Africa: The Sources of His Fiction. London: Hans Zell Publishers 105 SPEECH SURROGATES OF AFRICA: A STUDY OF THE FANTE MMENSUON Adwoa Arhine1 Abstract Various forms of communication based on sounds produced by instruments are common in many African societies. Among these, the slit gong and drums are the most popular and the most widely used as speech surrogates in Africa (Nketia, 1971:700). With the introduction of participant-observation research orientation to African scholarship, new information is being discovered and structured to fill existing gaps in knowledge. The ‘mmen esoun’ (also spelt ‘mmensoun’) (meaning Seven Horns) is a speech surrogate used among the Fante of Ghana. Mmensoun has the dual capability of imitating the speaking voice and, simultaneously, serving as a musical instrument in a performance. As a speech surrogate, it functions as an effective and powerful instrument for communication. This paper introduces the communicative sounds of the mmensuon within the cultural system of the Fante, and offers a framework within which the instrument could be further investigated. 1. Introduction The literature on instruments that imitate speech texts in Africa abound (Nketia, 2005; Avins et al. 1999; Creighton, 1999; Maxwell, 1999; Kofie, 1994). Among the Akan, Ewe, Mamprusi, and the Ga of Ghana, instruments such as the talking drums (atumpan), double bell, slit gong, trumpets (mmentia) and talking trumpets (ntahera or aseseben) are commonly used. In Congo, the bowed lute (sésé) serves also as talking instruments (Carrington, 1949: 79), while the hourglass drum is used most frequently among the Yoruba and areas of Dagbani ancestry. The Godie, a one-stringed fiddle among the Nigerians or Dahomey, assumes an almost human role in dance bands, and reproduces speech patterns and inflections note for note (Bebey, 1975: 44). The Masengo fiddle found in Ethopia is also a talking instrument (ibid, 44). In Niger, because of a system of phonetic equivalences, Djerma alghaita players “send messages … that can be decoded by the initiated” (ibid, 78). The Fang of South Cameroon also have the mendjang orchestra which consists of at least four or five xylophones; one of these xylophones played by the group leader ‘converses’ with the dancers. These instruments have had the benefit of more exposure to research, especially by scholars who popularised their findings. 1 Ms. Adwoa Arhine is a Lecturer in the Department of Music, School of Performing Arts, University of Ghana, Legon. 106 Legon Journal of the HUMANITIES Volume 20 (2009) Ethnographical research has shown that among the Fantes, there are other instruments that are assigned the talking attributes while the drums play percussive roles in the socio-cultural context. The traditional instrument chosen for this study is the mmensuon because it has the capabilities of imitating speech text or the speaking voice and simultaneously serves as a musical instrument in a performance. A mmensuon ensemble basically consists of seven horns; namely, one Sεsε, one far, two agyesoa, two ab so and one Otu. In the cultural context, the minimum number in an ensemble is four and the maximum ten. Apart from the Sεsε which ‘speaks’ clearly, all the others are supporting instruments and therefore they could be four or ten depending on the context of performance and what the players want to achieve; and the larger the number of horns in an ensemble the greater the volume of the sound and, vice versa. The groups in this study limit their instruments to the seven horns mentioned above. The mmensuon are traditionally considered sacred and mostly associated with the political hierarchy of the Fantes. The mmensuon is deeply rooted in the culture of the Fantes to the extent that its sound expressions are perceived as representations of deeply felt sentiments. This paper discusses the communication potential of the mmensuon sounds and how the interpretation of these sounds leads to an understanding of the socio- cultural life of the Fante people. The paper also sets a framework within which the instrument could be appreciated and further explored. 2. Background The Fante people constitute one of the ethnic divisions of the Akans. They can be found in the south western coastal region of Ghana (referred to as the Central Region), from Pra in the West to about 24km West of Accra, the current capital of Ghana. A few of them are also found in Yamoussoukro in La Côte d’lvoire. The traditional states which make up this ethnic group include the Assin, Fante, Asebu, Efutu, and Etsi with the Fante people exercising cultural and linguistic influence over the others. This process of cultural integration is still on-going since clusters of pure Efutu, Etsi and others are still found in the Central Region. It must be noted that the main Fante state comprises Mankessim, Abura, Ekumfi, Nkusukum, Ogua (Cape Coast) and Anomabo. 3. Methodology Data was collected using qualitative research instruments. Qualitative instruments used include, focus group discussion, key informant interviews and direct observation. Interview schedules were designed and used while note-taking, tape recording, video recording and photographs were used in recording responses from the respondents. Discussions were held with some men at the Cape Coast chief’s court Adwoa Arhine 107 and others in the town to elicit information on the history of the traditional horn ensemble among the Fantes and the intricacies involved in the practice and performance of mmensuon by the men horn ensemble in the cultural context. Apart from information obtained from Cape Coast, relevant information was gathered from the surrounding villages including Mankessim, Budu-Atta, Fawmanye, Gomoa Isaadze Eshrew and Otsew. All these villages have professional male mmensuon ensembles as well as similar Fante musical traditions. While interesting details of gender restrictions in some cultural activities may be the subject matter of another discussion, it is expedient at this point to explain that one of the major reasons for this behaviour is that some cultural items of the tradition are highly respected as religious icons. These items include major musical instruments such as the atumpan drum, etwie or snare drum, and mmensuon. The Atumpan (a double headed pair of drums), is played during ceremonial occasions by men. Etwie drum (snare drum) sounds like the leopard’s cry. This drum is played by men to announce a terrifying message such as war, or the death of a prominent person. From oral interviews, mmensuon is also the prerogative of men. The sεsε (lead instrument) is said to communicate with the living, ancestors and gods; therefore, the Fante people revere it to the extent that they see women in their menses as unclean to touch these reverent instruments. They believe women would contaminate the instruments and render them ineffective, and since one could not tell when a woman was in her menses, the best thing was to forbid them from touching the mmensuon. In view of this patriarchal bias, it was quite difficult initially to study the instrument especially in the chief’s court where tradition is highly rooted. Consequently, a physical touch of the mmensuon in the chief’s court was not possible but interviews and interactions were granted at the chief’s palace. Furthermore, I was granted permission to participate in some training and rehearsal sessions with okukurampon cultural troupe mmensuon ensemble at Mankessim. This group was made up of school boys and girls who were practising traditional music in the contemporary context, devoid of the aforementioned rigid cultural practices. Indeed, the playing of mmensuon was part of their cultural studies. The group gave me the opportunity to touch, experiment with, take photographs and ask questions about the instruments without any problems. I also observed public and private performances by both the boys and male groups. Some of the musical performances were also recorded and video taped uninterrupted at durbar grounds and at the chief’s court with a drum ensemble accompanying the horn sounds. Others were recorded during a staged performance on request. The performances were followed by focus group discussions with both the performers and a cross-section of the audience to ascertain the meanings and interpretations of the text. Most often the various answers were cross-checked with some other people from the community; or I related them to the physical and 108 Legon Journal of the HUMANITIES Volume 20 (2009) emotional behaviours observed during the performances. Consequently, rich data with regard to the sound text, context of performance, interpretation of concerned problems of aesthetic judgment, perception and values were assembled. 4. The Mmensuon Nomenclature The Fante language has grouped all air-blown instruments as mmen (plural) or aben (singular), which is very difficult to translate in English. The mmen could be trumpets, horns or flutes. They are practically air-blown instruments. Usually, it is the material used in making the instrument, the size or the number of instruments put together in an ensemble that identifies one group of mmen from the other; for example, abentsia (short horn/trumpet/flute), and mmensuon (seven horn/trumpet/ flute). This ambiguity has generated various definitions for the traditional horn and trumpet. Sachs (1940: 457) cited in Carter (1971) states that there is a clear- cut distinction between the horns and trumpets. According to Sachs, the horns are accurate reproductions of carved animal horns by their shape and have conical bores, whereas the trumpets are accurate reproductions of straight tubes of bamboo or branches of trees. Nketia (1958: 27) also makes a clear distinction between horns and trumpets. He indicates that horns are made of animal horns or tusks of elephants while trumpets are carved out of wood. Sibyl (1964: 244-247) defines the horn as a wind instrument initially made of animal horn or tusk of elephant and later replicated in wood or metal while trumpets are originally straight instruments made of wood, bamboo or cane. The above definitions do not confirm the word mmen as used by the Fantes in this context because the mmensuon were originally carved out of elephant tusk, but as time went on they were carved out of wood. Mensah (1966), an indigenous Fante, refers to the mmensuon ensemble at Gomoa, a trumpet set, while Philip Quarcoe Girls School educators who are mostly Fantes also call the mmensuon ensemble, a horn ensemble. Is mmensuon therefore a trumpet or horn ensemble? Mmensoun in this paper will refer to a set of seven horns. 5. Description Traditionally, Mmensuon (7 horns) are lip-vibrated, side-blown ivory horn instruments, in various sizes of small, medium and large depending on the size of the elephant. The small horns measure approximately 35 – 40 centimeters long while the medium and large are 45 and 50 – 55 centimeters long respectively. The bigger orifice of the mmen tapers to a smaller end where the opening is corked. However, the lead horn has a small hole through the cork, which is at the topmost part of the smaller end. There is another opening on the concave side of all the horns. These horns are distinguished by the names assigned them, which also reflect the role that they play in the ensemble. The smallest and high-pitched horn, Adwoa Arhine 109 which usually plays the lead or calls the tune has been given the name sεsε and is approximately 35cm long. This horn emits two notes, G and A. Fig 1. Each of the other horns emits only one note. Below are the names and sizes of the rest of the horns: Table 1. Name Centimeters Size Sound ffar 40 Small F Agyesoa and Ab so 45 Medium D Otu 50 Big Bb Fig 2. left: sεsε (lead), ab so / agyesoa, ffar, and otu. 110 Legon Journal of the HUMANITIES Volume 20 (2009) The sεsε’s larger counterpart, the abεso, emphasizes the sεsε’s announcing phrase with a special response; the far repeats the sεsε’s announcing phrase; the Agyesoa, which has great carrying power is regarded as the signalist and time-keeper among the group. It follows with a motif designed to heighten the general effect (Mensah, 1966:82). The largest, is called Otu. It has the deepest tone and is regarded as the eldest of the instruments. In the view of the performers, its utterances serve as a general endorsement on the statement being made by the rest of the set. Sometimes the biggest horn could be as long as 55cm long. From the above, it can be recognized that the difference between the three horns is 5cm. In other words, the length of the biggest horn minus the length of the medium size horn is 50 – 45 = 5cm; the length of the medium size horn minus the length of the small size horn is 45 – 40 = 5cm; and finally, the length of the small size horn minus the length of the smallest (sεsε) size horn is 40 – 35 = 5 cm. It can be observed that the length and size of the horn corresponds to the size of the embouchure; for instance, the smaller the horn, the smaller the embouchure. The size, length, and the embouchure accounts for the differences in the tones (pitch) of the horns. Below are the different horn sounds of the mmensuon ensemble and its harmonics, plus the areas of higher intensities expressed in decibel, being indicated by peaks of different contours and shades. A4 - 7 Cents Homonics Intensity Graphics In te ns ist y L ev el d B Adwoa Arhine 111 G4 - 14 Cents D4 + 11 Cents Homonics Intensity Graphics Intensity Graphics In te ns ist y L ev el d B In te ns ist y L ev el d B 112 Legon Journal of the HUMANITIES Volume 20 (2009) Bb3+ 70 Cents (Graphics: Courtesy P. Dominic) F4 + 21 Cents Homonics Intensity Graphics In te ns ist y L ev el d B The Homonics Intensity Graphics In te ns ist y i n dB Adwoa Arhine 113 The beginning, modulation as well as the diminishing process of the sound in each graph are different and most of the times, they continue quite irregularly. In the case of the A and G, the 1st and 2nd partials build up rather quickly, while the upper partials need more time to build up their full energy; in the case of F, D and Bb, the first two partials build up more slowly than the upper partials. This is the reason why the sεsε’s parts sound more clearly defined from the beginning than the supporting mmensuon. In all the graphs above, the higher harmonics are relatively diminished because with mmensuon blowing, the pressure of air gathered in the lungs and released accounts for the periodic wave, which consists of a fundamental and its harmonics. When the performer blows hard into the horn, the vibrations increase and the fundamental and its entire harmonics also rise in pitch and sound, and afterwards die out. The reverse is the case. The researcher believes that comparatively, the pressure of air gathered and released for the G sound was more intense than that of the A sound taking into consideration partials 9 to 15 of the A sound and that of the G sound which is full of peak areas. These contours indicate that for low tones the pitch decreases with intensity, but for high tones, the pitch increases with intensity. The intensity of the fundamental and particular partials affects the quality of the sound. This is in line with Lundin, (1953:49) who states that when a musician varies the intensity of his/her produced tone, s/he varies its quality also because the greater the intensity, the more partials will be present. Table 1. Peak or high intensity areas of the various sounds of the mmensuon; Graph A4-7 cents 2 4 6 8 11 14 Graph G4-14 cents 2 4 5 8 10 11 14 Graph D4+11cents 1 2 4 6 8 11 13 15 Graph F4+21 cents 1 3 4 6 8 10 14 Graph Bb3+70 cents 1 3 4 5 8 11 13 15 The figures in the boxes are the overtones of the various sounds. Overtones 4 and 8 are peak areas of all the horn sounds while overtone 1 is a peak area for D, F and Bb. Overtone 2 is also a peak area for sound A, G, and D. This implies that around these peak areas, the sound of the horn is very intense or loud in intensity. With reference to the harmonic series, 2nd, 4th, 8th and 16th are octaves. Therefore, it may be opined that due to the hollow shapes of the mmensuon, which allows the opening at one end to reflect the higher overtones and resonance when the lips are applied with the appropriate techniques, playing in octaves is the easiest and the most pronounced sound quality one could achieve in mmensuon sounds except the 114 Legon Journal of the HUMANITIES Volume 20 (2009) 16th partial which cannot be perceived. People who cannot stand loud sounds are usually affected at these levels where the sound is more pronounced. Sometimes unconsciously, some people get startled whilst those indigenous Fantes who are familiar with such intensities rather show signs of elation. On the contrary, blowing softly may not be as easy as blowing loudly. The table below shows the soft or low intensity sound levels: Table 2 Graph A4-7 cents 1 3 5 7 9 10 15 Graph G4-14 cents 1 3 6 7 9 13 15 Graph D4+11 cents 3 5 7 10 12 14 Graph F4+21 cents 2 5 7 9 11 15 Graph Bb3+70 cents 2 6 7 9 10 14 Again with reference to the harmonic series, 5th, 3rd, 2nd and the 7th harmonics are weak and therefore the intensity levels are low. This might explain why the mmensuon plays in hocket technique so that by alternating the notes, the various sounds would be heard properly because the attack points will vary. This hocket technique immediately creates a community in the relationship between the sounds as also practised by such instruments as the mbira. 6. Playing Technique and Sound Production The traditional horn is played in transverse position with the thumb placed at the smaller end while air is blown through the embouchure. A significant discovery was made when the researcher was learning how to blow into the sεsε. The carver, who is also a Fante used some specific words in teaching how to blow into sεsε – Ka w’ano b do si tokur n’ano, na huu mframa gum (translated as, Make your lips tight, place them on the embouchure and blow hard into it). Many players position their mouth such that it covers more of the upper lip than the lower lip. The loudness of the sound that is a psychological sensation depends somehow on the mouth pressure and the manipulation of the thumb of the player. The harder s/he blows into the mmensuon and opens the thumb, the louder the sound produced and vice versa. 7. Training of Traditional mmensuon performers Research findings in this study revealed that, among the Fantes, traditional music is learnt through enculturation. Looking at the techniques in blowing the horns, Adwoa Arhine 115 one would think that mmensuon playing requires special training, but, the learning of mmensuon is not different from the learning of other traditional instruments. Nketia (1964:4) describes how traditional musicians learn through slow absorption without formal teaching. He states that the individual is required to acquire his musical knowledge in slow stages, to widen his experience of the music of his culture through the social groups into which he is progressively incorporated and the activities in which he takes part; the young have to rely largely on their imitative ability and correction by those who this is volunteered. They must rely on their own eyes, ears and memory. They must acquire their own technique of learning. Davis (1994:27) also states “you have to use your common sense right there to make sure that you get the patterns clean”. This is exactly how the novices of Kofi Ninsen’s professional mmensuon at Gomoa and other professionals learn the techniques of playing. Among the Fantes, the traditional belief is that a genius youngster may be the reincarnation of an ancestor who was a renowned musician. Although it is taken for granted that natural endowment and a person’s ability to develop on his/ her own are the requisite qualifications for becoming a mmensuon blower, how the skill is acquired is a personal responsibility. The basic principle therefore remains that learning is through social experience, which is by assimilation, participation and imitation. This is based on the African worldview “that you do not teach a blacksmith’s son his father’s trade” (Nketia, 1974). Under the prevailing training tradition, the candidate’s chance of mastering a piece takes a very long time. Due to the important role the mmensuon plays in society, it has survived many generations among the Fantes and has not been transformed very much. In this contemporary era, mmensuon is still practised in the chief’s courts in the Fante community. The greatest innovation is the fact that currently mmensuon has moved from the traditional courthouses and durbar grounds to institutions, theatre halls, state houses, churches and concert halls. 8. The Mmensuon performer in Fante Community Traditional mmensuon blowers in Fanteland are of two categories; the ones attached to the courts and those associated with a particular patron. The court musicians are highly respected because of their connection to royalty. Most often a particular lineage is chosen to perform such duties. This family lives in the chief’s palace at the expense of the chief. He feeds and clothes them as well as gives them some honorarium because their livelihood is dependent on playing music for the chief. Since they spend all their lives in the court, they are familiar with the histories of the land, the norms and 116 Legon Journal of the HUMANITIES Volume 20 (2009) traditions governing court music making, choice of songs for each state function and many other conventions associated with court-music performance. This knowledge is therefore passed on from one generation to another. As they live in the same courtyard, they meet on several occasions to rehearse and build rapport among themselves over a long period of time. This makes them more skilful and therefore they perform with confidence and dexterity. Nketia (1963) describes similar practices concerning the Ashanti court musician. The second category of mmensuon performers are individuals brought together and trained to play for commercial purposes. They move every now and then from one village to another performing during funerals and durbars. Some of these groups are found in Isherew and Budu Atta, villages in the Mankessim area. Relatively, the court mmensuon performers are more skilful than the commercial performers because their experience starts from childhood through observation, imitation, creativity and in-built talent; while most commercial performers acquire knowledge at a later stage and with time leave the group for a new set of people to start all over. 9. Mmensuon Song Text and Social Sentiments Oral tradition revealed during the field investigations (January, 2002) that the mmensuon plays a dual role in the cultural context. As an ensemble, mmensuon is a regular feature in the typical traditional court that performs in contextual situations such as traditional festivals, funerals, durbars and anniversaries. Apart from its musical capability, the lead horn (sεsε) is used in the court as a talking instrument for recounting histories, singing appellations, uttering proverbs, and conveying messages, announcements and signals depending on the context. The sεsε’s textual content is highly idiomatic and proverbial, just as the Fante language it imitates. It takes an insider (in this case the indigenous people) to decode these expressions since the meanings of the horn sound are governed by agreed principles within the tradition; therefore, people outside the tradition would find it difficult to interpret and understand them. Wherever it features in any occasion, ceremonial or occasional, the mmensuon’s role is that of a speech surrogate as well as a vehicle to convey the social sentiments of the Fantes. The first example is illustrated by the following text of the sεsε’s trumpeting during the burial of a prominent chief at Mankessim as observed. The sεsε leads the way to the cemetery as it trumpets: Duei oo o Sorry oo o Nana brεbrε Chief slowly. Adwoa Arhine 117 The interpretation given was that culturally chiefs are not supposed to walk briskly. They do things majestically when they are alive; therefore, when they die the same sentiment is expressed to even the dead body. A similar performance takes place at other Fante courthouses during funeral occasions as told by the entire major informants. Another example can be mentioned of Pata, a dirge at the Borbor Abora’s chief’s court which has the text as follows; yε a wonsu mu Hold it well Ma nnka famu So it does not touch the ground Odε me pata rebuo He says my hut is breaking yε a wonsu mu Hold it well Ma nnka famu ara o So it does not touch the ground yε a wonsu mu o Hold it well Ma nnka famu oo So it does not touch the ground Among the Fantes, it is a taboo for a chief to fall down whether dead or alive. The chief is very important to the community, and that feeling of affection is expressed through the mmensuon song text even to the dead body. At the B rb r Abora chief’s court, the mmensuon plays appellations with some provocative implications when the chief or queen mother is going to a durbar ground for example, hen na rek no, brεbrε The chief is going, slowly hembaa na rek no, brεbrε The queen mother is going, slowly Kwansaba Kweku Kwansaba Kweku ei hembaa na rek no, brεbrε The queen mother is going, slowly hen na rek no, brεbrε The chief is going, slowly There are two versions of interpretation to the text above. It reflects two different social sentiments about dress codes in the society. The interpretation given at Cape-Coast was that when the sub chiefs are attending a durbar, each one wants to look his or her best even though they all appear in their traditional Kente clothes, which is the norm. Among the chiefs, dressing appropriately then is a competition. This song implied that a particular chief’s dressing befitted the office; thus, the praise “there goes a chief or queen”. The text was at the same time interpreted as mocking the other chiefs who were not as elegantly dressed as the one being praised. The point at issue is not the correctness of the interpretation of a particular version, but rather the nature of the social sentiments expressed in each version. 118 Legon Journal of the HUMANITIES Volume 20 (2009) Mmensuon ensemble also imitates the text below, which has some provocative implications. Kofi Ninsen’s mmensuon group, which is attached to the Gomoa Fawowmaye’s chief’s court as well as a professional group that moves from one village to another, performed this song at a durbar. yε wo asεm ben? Is it your palaver? fa wo ho ben? Does it concern you? M’edzi me Nana n’adze a If I’ve inherited my grandparent yε wo asεm ben? Is it your palaver? fa wo ho ben? Does it concern you? Most often, among the Akans, chiefs are selected from a particular lineage just as mmensuon performers are chosen from a particular lineage. Therefore, the stool is passed on from one generation to another. However, when there is a polygamous marriage at a particular point in time of a chief’s tenure, confusion arises as to which of the two wives’ son is to inherit the throne after the death of their uncle. In the midst of such inheritance controversies, jealousies and hatred occur among the wives and children. This situation also sometimes happens to any other person in the community who has inherited some property that is shrouded in controversy. This text expresses social tendencies such as hatred, anger and jealousies of this nature. Festivals and traditional customs also have songs that express the feeling of joy and happiness. When the chief is carried in the palanquin dancing on the streets of Cape Coast during festivals and other durbars, the procession sing and dance to the music being performed by the various musical groups. The mmensuon also sing eulogies as well as express the social sentiments that the people share as they celebrate the festival. Nana Krampa b k Chief Krampa, slowly. Yee yee b k Okay, okay slowly. On the other hand, mmensuon song texts also serve as a means of advice to the community people. A typical example is a song text by the Gomoa Budu-Atta, mmensuon ensemble that goes as follows. Me da me dabi ara a I sleep my style of sleep Mewu nda mu I die in sleep Kofi Ninsin e! Kofi Ninsen! Meda me dabi e! I sleep my sleep! Mewu nda mu a I die in sleep Abowa nanka Reptile serpent Adwoa Arhine 119 The interpretation given to this song by Opanyin Kweku Edu of the chief’s court at Budu-Atta was that God made man to work before he could get a means of living. If one decides not to work but make it a habit to lazy about or sleep a lot, there is no way that person could get a means of survival. Therefore, everybody should rise up and work. Usually, when the ensemble is performing some musical pieces in the area of concern such as implicit provocation as the one above, the actual name of the person implied is not mentioned; however, the name of any of the performers is mentioned instead. In the example given above, the name of the leader of the group that is Kofi Ninsen was used instead. Another example of such implicit mmensuon song is the one below: Gyang s , enye woana b g r? Old woman who will you play with? Ihu fie bowa a erasa h n When you see a family, you dislike the person. Ihu ham bowa a eresar h n When you see a visitor, you dislike the person. Gyang s , enye woana b g r? Old woman who will you play with? This is a story of an old woman who was picking quarrels with anybody who came into contact with her and therefore she was not on talking terms with almost all the people around her. Then one day the inevitable happened and she fell ill. Who will care for her needs? The underlying issue is to be nice to everybody and live at peace with your neighbours. Outside the court, the mmensuon serves purely as a means of entertainment. They perform at funerals to console the bereaved families, which, most often ends up evoking tears in the eyes of those who are assembled. This affection that could be viewed on the level of aesthetic consciousness is partly due to the text of the pieces performed and greatly due to the tone of the instrument at that particular time which was able to depict an atmosphere of sorrow. The mmensuon texts effectively communicate the deeply felt sentiments that are embraced in the sociocultural setup of the Fantes. Due to the important role the mmensuon plays in society, it has survived many generations among the Fantes and has not been transformed very much. In this contemporary era, mmensuon is still practised in the chief’s courts in the Fante community. The greatest innovation is the fact that currently mmensuon has moved from the traditional courthouses and durbar grounds to institutions, theatre halls, state houses, churches and concert halls. 10. Summary, Conclusion and Recommendation The double role of the mmensuon signifies the use of mmensuon as a communication tool to express various forms of social sentiments and as a musical instrument for ceremonial purposes and currently for recreation. 120 Legon Journal of the HUMANITIES Volume 20 (2009) Horn sound also depicts what the Fantes know from their observational and practical experiences during their journey to the current habitat through the tribal wars they fought both successful and failed ones, their interaction with their neighbours, as well as deeply rooted cultural values. Since mmensuon is associated with the courts and was the preserve of brave and “holy” men, many of the themes reflect the historical origins, ancestors, wars, appellations, social vices, messages or signals, beliefs and eulogies. The article revealed that Fantes conceptualize mmensuon sounds as speech surrogates and therefore the sounds are conventionally associated with the language. They are composed with deliberate goal in mind. For example in the context of death, the horn sound is organized deliberately to evoke that mood. This explains why the mmensuon sounds are powerful tools for the communication of sentiments. Mmensuon texts are also a means through which Fantes communicate with the purpose of controlling social vices. The totality of mmensuon performance is the core of Fante aesthetics because it has the unique power to evoke and express social sentiments. Mmensuon song texts can therefore serve as a solid foundation for sociocultural studies in the Central Region, and appropriated in other civic educational contexts. Adwoa Arhine 121 REFERENCES Arthur, Simon 1989. Trumpet and Flute Ensembles of the Berta People in the Sudan. In African Musicology, Current Trends Vol.1. Jacqueline Cogdell DjeDje & William G. Carter (ed.). Arhine, Adwoa 2004. Sound and Sentiment: An Interpretation of structures in FanteWomen Horn Ensemble. A case of Philip Quarcoe Girls School, Cape Coast. M.Phil Thesis, University of Ghana, Legon. Unpublished. Avins Lyn et al 1999. The Heritage of African Music. Cameron E.L et al. Regents of the University of California. Carter Williams G 1971. The Ntahera Horn Ensemble of the Dwaben Court: An Ashanti Surrogating Medium. M.A Thesis, University of California. Los Angeles. Chernoff, John Miller 1979. African Rhythm and Sensibility. Aesthetics and Social Action in African Musical Idioms. The University of Chicago Press- Chicago. Carrington, J.F. 1949. Talking Drums of Africa. London, Harry Kingsgate Press. Feld Steven 1982. Sound and Sentiment. Birds, Weeping, Poetics, and Song in Kaluli Expression, Philadelphia Press. Heather A. Maxwell 1999. West Africa: When the Xylophone Speaks. In Turn Up the Volume! A Celebration of African Music .Cogdell DjeDje Ed et al. Regents of the University of California. Leigh Creighton 1999. The Luna Drum as a Social Mediator among the Dagbamba of Ghana. In Turn Up the Volumn! Cogdell DjeDje Ed et al. Regents of the University of California. Kokroko, J.A. 1990. Akan Instrumental symbolis. Diploma Thesis, University of Ghana Mensah, A.A. 1966. the Gomoa OtsewTrumpet Set. In Research review Vol.3 No. 1. Institute of African Studies. University of Ghana Michaelmas 122 Legon Journal of the HUMANITIES Volume 20 (2009) Term. Mokwunyei, O.N. Josephine 1999-2001. A Study of the Aniocha-Igbo Trumpet. Legon Journal of the Humanities. Vol. 12. Faculty of Arts, University of Ghana. Nketia, J.H. Kwabena 1971. Surrogate Languages of Africa. Thomas A Sebeok (Ed). Current Trends in Linguistics in Sub-Saharan Africa, Vol.VII, pp. 699-732. The Hugue, Monton. ………………….. 2006. Sound instruments of Ghana. Afram Publications (Ghana) Ltd. ………………….. 2005. Modes of Inquiry & Interpretation Vol. 1.Ethnomusicology and African Music (Collected Papers). Afram Publications (Ghana) Ltd. ---------------------- 1982. The Musical Traditions of Akan. Tarikh No.26. Akan History &Culture 7 (2): 47-9. ------------------------ 1963. Drumming in Akan Communities of Ghana. Edinburge, Thomas Nelson and Sons. Okyere, Vincent N. 2000. Ghana: A Historical Survey. Vinojab Publications, Accra. Tsukada, Eenichi 2001. Asafo and Fromtomfrom: conflict and Unity in Fante society of Ghana. In cultures Sonores D’Afrique II- Aspects-Dynamiques Public sous la direction de Kawada Junzo et Tsukada Kenichi. Hiroshima City University, Japan P.33. 123 The Phonology-Syntax interface in Avikam Firmin Ahoua1 Abstract Very often African languages have been claimed to have tones but no intonation which may be understood as some organization of complex prosodic or breath grouping, beyond the segment. However, recent studies have shown that these languages may have complex prosodic structures. Avikam, an isolate lagoon language of the Kwa cluster spoken in the Ivory Coast, is investigated. The paper argues that Tone Lowering is a well established phonological rule that is constrained both by syntactic and prosodic domains, in a quite intricate way. This interaction points to an interface level between syntax and phonology, which may differ from a mere postlexical component. Based on the Tone Lowering rule, prosodic domains such as the phonological word, the phonological phrase and the tonal foot are identified. The paper provides evidence that recursion of the prosodic word is a common process similar to that attested in Leben and Ahoua (1997). The paper has implications for the strong version of the Strict layer Hypothesis (Selkirk 1978), for which some weakening may be required. 1. Introduction The goal of the paper is to describe Avikam complex tonal rules in nominal and verbal constructions and examine their interactions with the prosodic, syntactic and segmental domains and constraints. The interactions between syntactic constituents and phonological constraints provide interface structures that current theories call “the prosodic domains”. The latter presumably affect phonological rules and “readjust” the syntactic structures before the phonetic implementation applies (cf. Truckenbrodt (1995, 1999), Leben and Ahoua (1997), Inkelas and Zec (1990), Nespor and Vogel (1986) and Selkirk (1978)). The paper argues that Tone Lowering is a widely attested phonological constraint in Avikam that may contribute to motivate the distinction between the prosodic word and the phonological phrase with respect to syntactic domains. The prosodic word is argued to be required for the analysis of prefix apheresis and unbounded Low tone spreading. The paper accounts for some exceptions 1 Professor Firmin Ahoua is a Professor of Linguistics in the University of Cocody, the Ivory Coast. 124 Legon Journal of the HUMANITIES Volume 20 (2009) to the Tone Lowering Rule generalization by adducing independent evidence from sequences of monosyllabic words. The prosodic word is claimed to have recursive properties. The prosodic phrase on the other hand generally correlates with clauses, and sometimes violate for instance Noun and Adjective and Noun and numeral phrases. The prosodic phrase helps to account for some tonal asymetries between double object constructions and possessive constructions. I then provide evidence that the unbounded Low Tone spreading in disyllabic words is constrained by the tonal foot (cf. Akin and Urua, and Urua 2007 for motivations of the foot ). At the syllable level, the Tone Lowering Rule helps to decide whether the surface [CGV] syllables are to be analysed as diphthongs or as a sequence of vowels (disyllabic words) (cf. Ahoua and Leben 1999). The paper thus attempts to contribute to the typology of the prosodic domains and hierachy and their interaction with syntax in Kwa languages, using the recursive approach by Leben and Ahoua (1997). The paper is organised as follows: Firstly, I provide an overview of the location and classification of the language; secondly, I survey the vowel and consonantal system as the latter is involved in the tonal rule conspiracies of the language. The last sections discuss the interaction between tonal rules and domains. Location and classification of Avikam2. Avikam is a Kwa language of Côte d’Ivoire located between 5° N. and 6° N. Latitude and 4° and 6° W. Longitude. Genetically the language belongs to the Volta-Congo, a subgroup of Niger Congo languages. Stewart (1989) classified the language as an isolated lagoon language under the Nyo cluster, together with Alladian to which it is closely related. The number of speakers is estimated at about 30,000 speakers spread in thirty villages. The language is spoken along the coastal plain of Grand Lahou and Toukouzou and is bordered in the North by Didas, in the West by Godies, in the East by Ahizis and Alladians and in the South by the Atlantic ocean. A careful investigation of the language together with a group of native speakers who have requested us to prepare a literacy primer for them has revealed at least two different dialects. Apart from the mutual intelligibillity, the phonological differences are minor and seem to be limited to segmental correspondences between labial obstruents and labial glides (cf. The word for ‘tree’ [eziba] vs [ezuwa]), alveolar fricatives with palatal ones (cf. [besi] ‘plantain bananas’vs [beʃi]) in some dialects. I also noted the replacement of final high front vowels with high rounded vowels, as well as a few lexical variations. The language has been little described. Only few major linguistic works have been published on aspects of its phonology and grammar (Kadio- Firmin Ahoua 125 Morokro (1978), Duponchel (1971), Herault (1982), Rongier (2002)). None of the preceding works has motivated the Tonal Lowering rule in the nominal system, and its complex interactions with other prosodic domains has not been noticed, though valuable work by Herault (1982) provides some brief remarks on some tonal effects in the verbal system. The present work is the result of several field works in the area of Grand Lahou on the Kpanda dialect throughout the year 1999. The data has been checked together with many informants to ensure the maximum of consistency. It reflects the present stage of the tonal system of the Kpanda dialect. Here is the opportunity of calling the attention of researchers on the risks of working with only one informant because we encountered variations that were often rejected by other native speakers. We also often met disagreements on some details of pronunciation. Tone is the most sensitive area where individual variations may occur. In that respect, only contrastive pairs taken from the lexical inventory of the language and which have been proved to be stable and documented in various distributions were used as reliable heuristic tools. The segmental system of Avikam3. The Consonants3.1 The language has voiced and voiceless series of sonorants and obstruents including lateral and labial approximants. [ʒ] is a contextual variant of /s/ when followed by a front high vowel. Avikam has only one labial implosive that Hérault (1982) classifies as a sonorant along with the approximants [l],[j] and [w], an analysis that we adopt in our Figure 1. Its pronunciation is auditively close to that of [m] (cf. Ladefoged, 1969: 6) for a similar observation on Ebrie). There is, however, no synchronic alternation between the nasals and the sonorants. Following Herault (1982) who has set up a full class of sonorants by including the series of glides, the present classification is represented as in Figure 1 and is motivated later on the basis of the tonal phonology, an argumentation that is not based on the need to satisfy the principle of pattern congruity and symmetry. 126 Legon Journal of the HUMANITIES Volume 20 (2009) Labials alveolars palatals velars labiovelars stops: voiceless p t C k kp stops: voiced b d ɟ g gb Sonorants ɓ l J Ɣ w Nasals [m] [n] [ɲ] [ŋ] Fric: voiceless f s Fric: voiced v z [ Ʒ] FIGURE 1: The consonant system of Avikam The segments represented in Figure 1 are all contrastive except the ones in brackets. According to Hérault (1982) all the nasal consonants are distributional variants of the approximants. A similar case has been postulated for Potou languages. The status of [m] is less obvious as it may be analysed as a variant of its stop /b/ or sonorant / /counterpart. Synchronically, there exist environments where the implosive is realised as nasal [m]. However, there exists also a true underlying /m/ as the tonal alternations will show. The latter is opaque to tonal spreading while the surface [m] that is derived from / / is transparent to the same tonal process. Regarding the consonant / /, it should be noted that Avikam belongs to the few Kwa languages that have maintained some older phonological forms of proto- Kwa (cf. Ahoua (2006) for similar data in Tano, a Kwa subgroup). 3. 2. The vowels According to Hérault (1982:283), Avikam seems to belong to the conservative types of languages of the Kwa group that has the full set of advanced tongue root (ATR) and non advanced tongue root (-ATR) patterning in lexical items. However, it is crucial to point out that Hérault (1982:262) admits having found no contrast of words containing the (ATR) vowels. Lexically, Hérault (1982) observed that retracted High vowels were attested only in some twenty words and in borrowings: ”...la langue n’offre pas d’exemples d’opposition des rétractées [I] et [ ] ni entre elles ni aux autres. L’antérieure [I] est attestée dans une vingtaine de mots où elle n’est jamais initiale : emprunts parfois (le plus souvent à l’anglais : bàgɪ, sac), mots autochtones le plus souvent Firmin Ahoua 127 (bɔlɪ,rosée, asɪsɛ, joue,kɪka, où?, etc...). [ʊ] n’a été relevé que dans une dizaine de mots (...)” Following Herault (1982) and Duponchel (1971), we represent the vowel system as follows: +ATR -ATR +ATR -ATR oral nas. Oral nas. Oral nas. Oral nas. Oral Nas. High i ḭ ɪ u ṵ ʊ Mid e ɛ ɛ̰ o ɔ ɔ̰ Low a a̰ FIGURE 2 : The vowel system of Avikam In general, the ATR alternations of the type that is attested in most Tano languages are quite limited in Avikam to the domain of Noun phrases (determiners and nouns), and verbal syntagmas (verbs and their clitic pronouns). In addition to Hérault´s observation, Avikam uses only some odd thirty archaic words with the -ATR high vowels. Regarding nasality, the language has nasal vowels that contrast with oral vowels. The vowels [e] and [o] never occur following a nasal consonant. 4. The tonal system of Avikam 4.1 The syllable structure Avikam has at the surface only words with open syllables and maximally two consonants at the onsets. The inventory of the syllables contained in words are: V, CV, CCV, CVV that also occur in longer words. Avikam is a language with prefixes attached to words. Monosyllabic CV words are thus rare in citation forms. In contexts, however, the prefixes may be dropped under certain prosodic and syntactic conditions . In our analysis, whenever VCV words drop their prefixes in context, they would behave as monosyllabic CV units . The syllables of the type CVV will be phonologically analyzed as vowel sequences because the tonal assimilation rule may affect only one of the two vowels in the CVV words. Gliding of the first vowel in such sequences is argued to be just a surface phenomenon (cf. Ahoua and Leben 1999). 128 Legon Journal of the HUMANITIES Volume 20 (2009) 4.2 Lexical contrasts Avikam has two contrastive lexical tones, High and Low. It has also a downstepped High tone. There are no contrastive lexical Mid tones. The language has no contrastive contour tones on monosyllables, though it has, according to Hérault (1982:267), High-Low and Low-High contours on some syllables in initial, medial and final positions. In verbs, contour tones are clearly derived tones. A short glance at the lexicon (as for instance given in Hérault´s (1982) Atlas, volume 2) provides minimal pairs of level tones. As mentioned above, CV words don’t occur in isolation, so the contrastive pairs have all a prefix vowel. Consider the following examples : (1) Low-High High-High a. ɛs̀ ‘tomorrow’ ɛs̀ ‘fish’ b. ɛǹ ̰ ‘dance’ ɛǹ ‘mouth’ c. ɛɲ̀ ‘animal’ ɛɲ̀ ‘flesh’ It should be noted that a High in a Low High sequence as in (1) is not automatically downstepped. The Downstep can be analysed as a floating tone that lowers following High tones but its position is not predictable, though it generally occurs word internally as in (2) : Downstepped High v. High and Low: (2) Downstepped High High-Low a. zr ǃɓ ‘tongue’ ézîɓà ‘tree’ b. w ǃɓ ‘eye’ éŋglúbà ‘dance’ c. éklóǃvlî ‘head-luggage support’ áwólógà ‘animal’ d. ébróǃbú ‘coconut’ �trád̰à ‘field’ As we can see in (2) a sequence of H H!H contrasts with a sequence of H H L, where the last High is respectively realized as a Mid in the left row and as a Low in the right row. Across lexical words, tones distinguish sentences such as in (3) and (4): Firmin Ahoua 129 (3) a. èflúɓà ‘a book’ b. ɛ:̀flùbà ‘his book’ (4) a. mà ná ̰làvrí sɛ̰̀ ‘I give Lavri’s fish’ b. mà naá ̰làvrí ɛs̀ ̰ ‘I give Lavri a fish’ One question that may now arise is whether the tones in (3)b. and (4)b. are derived tones and whether they are triggered by a Tone Lowering Rule. The issue is discussed in greater detail in the next section. Let´s mention that the contrasts in (3) and (4) have been recognized by Hérault (1982:269) who however fails to generalize the rule to the whole tonal system and to other components of the language : ”Dans ce type de syntagme, le déterminé prend le plus souvent une tonalité syntagmatique basse : tel est ici le cas avec èfrú á.” Hérault (1982:269) 4.3 Tone Lowering and High tones neutralisation The Tone Lowering Rule in Avikam is mainly, but not uniquely, motivated in possessive or associative constructions (cf. Williamson 1986) that involve either a possessor noun and a possessed object or a possessive pronoun and its object. It is significant to note that the Lowering rule neutralizes the phonetic contrast between a disyllabic lexical word that has respectively a Low and High Tone and a disyllabic lexical word with respectively a High and High Tone. At this point, one may be tempted to informally propose the rule as follows: (5) Tone Lowering: H  L / X] possessor [___ We intend to argue below that rule (5) will be misleading and that the hypothesis of proposing a floating tone better accounts for the associative or possessive constructions, in addition to the fact that we have observed that a floating tone independently exists word internally as in (2) above. 130 Legon Journal of the HUMANITIES Volume 20 (2009) Lowering of monosyllabic words in possessive constructions1.1.1 Monosyllabic words occur only as non initial constructions and are derived from VCV lexical items that lose the initial vowel.The lowering of monosyllabic words is best motivated by lexical VCV items that have the High-High tones as the ones with Low-High tones. The examples (6) and (7) illustrate the neutralisation of the underlying High tone, making it phonetically identical to a Low tone. In the following examples ‘Lavri’, a proper name, is constructed with respectively a lexical High-High possessed nouns versus a lexical Low-High possessed noun. (6) Low-High tones Possessive Construction ɛŋ̀ṵ làvrí ŋṵ̀ Lavri’s hair ɛb̀ɔ̰ làvrí bɔ̰̀ Lavri’s hand ɛk̀á làvrí kà Lavri’s place ɛg̀bɛ̰ làvrí gbɛ̰̀ Lavri’s money In (6) we may be tempted to assume that the deleted lexical Low tone of the prefix is the trigger of the lowering but , as we can see in (7) below, even words that had prefixes with lexical High tones are lowered. (7) VCV High-High tones in isolation Possessive construction with prefix deletion �v� làvrí vɛ̀ Lavri’s medicine écú làvrí cù Lavri’s sea �sɛ̰̀ làvrí sɛ̰̀ Lavri’s fish �sɔ̰̀ làvrí sɔ̰̀ Lavri’s house �dɔ̰̀ làvrí dɔ̰̀ Lavri’s village �và làvrí và Lavri’s court átí làvrí tì Lavri’s buttock Although CCV words may pattern differently from CV words as attested in Kwa languages such as Baule or Krobu, Avikam CCV words follow the regular process as the CVs. Note that Adiukru, a kwa language geographically close to Avikam, CCV syllables are barriers to tonal rules as opposed to CVs (cf. Ahoua and Leben 2006). Firmin Ahoua 131 (8) VCCV High-High tones Possessive construction �zrà làvrízrà Lavri’s language �fl� làvrí flɛ̀ Lavri’s okra �brá làvrí brà Lavri’s wooden vessel In all these examples, the High tone of the possessed CCV nouns is realized as a Low tone. Lowering of disyllabic and trisyllabic words1.1.2 Within words containing two or three syllables, Tone lowering applies only on the first syllable starting from the left edge. (9) VCV High- High tones Possessive construction lówú làvrí lòwú Lavri’s bone sáká làvrí sàká Lavri’s rice dám làvrí dàm Lavri’s cigarette ɓás bá làvrí ɓàs b̰á Lavri’s boy naɟ̰óbá làvrí nàɟ̰óbá Lavri’s bird ɓásál� làvrí ɓásál� Lavri’s wife císál� làvrí císál� Lavri’s cloth ɓ�k�m làvrí ɓ�k�m Lavri’s navel The rule is straightforward. It applies if a High tone or a sequence of High tones follows a lexical phrase. In an associative or possessive construction, the first High tones become Low, depending on the prosodic structure of the word. Now, we are in a better position to choose to postulate a floating Low tone, let us call it, the associative tonal morpheme that may represent a morpheme of the associative marker, a solution that has been often proved useful in Niger-Congo languages (cf. Welmers 1963). The Low tone can be interpreted as ‘docking’ on the first syllable to its right. In the following section we intend to show that a possessive pronoun 132 Legon Journal of the HUMANITIES Volume 20 (2009) may exist between the possessor and the possessed that is not phonetically realized. At this point, Avikam resembles many other Kwa languages in which a floating low tone occurs in associative constructions and in predicate (subject-verb) constructions, as has been noted in Anyi, Nzema and Akan (Dolphyne 1986). 4.3.3. Lowering of words in possessive or associative pronouns and compound nouns The regularity of the Tone lowering process extends to the constructions with a proclitic associative pronoun and a noun . Consider examples with ‘sáká’’rice’with only High tones: (10) m̀ m̀ sàká ‘my rice’ À à sàká ‘your rice’ è or ɛ:̀ ɛ:̀ sàká ‘his rice’ ò or ɔ̀ ɔ ̀sàká ‘our rice’ ɔ̀ ̰or w̰ ɔ̰ ̀sàká ‘your rice’ ɲɔ̰ ɲɔ ̰sàkà ‘their rice’ In (10) the first column represents the citation variants of the pronouns depending on the vowel features of the following noun. Notice that the first High tone of the noun ‘sáká’ has changed to Low. Following Ahoua (1986), one may not exclude that the associative pronoun may be the one that is inserted between the proper nouns and the associated nouns, but which is deleted segmentally at the surface phonetic level. Some exceptions to the regular process of High tone lowering are given in (11) : (11) �dá làvrí dá Lavri’s bundle of twigs égbó làvrí gbó Lavri’s shell égbó kébé gbó crab’s shell �ɲ ̰ �sɛɲ̰ ̰ fish meat édí ɛt̀ì� dí dog’s excrements (Hérault 269) �kl ̰ làvrí kl Lavri’s hedge ékpó làvrí kpó Lavri’s half part ékrw� làvrí krw� Lavri is not mature A question that emerges now is whether a floating tone occurs between these words or whether it is part of these items that blocks the Low tone docking. Indeed, an Firmin Ahoua 133 example such as the last one may be interpreted as a predicate phrase because High tones in adjectives are never lowered after verbs (in contrast they lower if they are part of the noun in a noun phrase). However, the remaining examples in (11) would resist this analysis. A solution towards the alienable or inalienable seems straightforward to maintain for the very first example: ‘bundle of twigs’as well as for ‘hedge’. This would also suggest that it´s not accidental that all the other exceptions are to be interpreted as inalienable cases. Another alternative would be to suggest a morphophonological analysis that requires a less abstract mechanism and involves phonological effects triggered by domain -edges or the labelling of syntactic categories. At this point, neither a morphophonological or morphological approach can handle the cases in (11) unless we assume a prosodic phrase that is independently motivated in other parts of the grammar (see Section 5). 4.3.4. Lowering of monosyllabic words in postpositional phrases In constructions involving nouns and postpositiona phrases, Tone Lowering also applies. Here again, a morphological analysis that would require an associative tonal morpheme would never surface in such constructions. (12) VCV High-High tones Noun+postposition construction �nám̀ làvrí nàm̀ behind Lavri ávázù làvrí vàzù before Lavri �n�̰sɛ̰̀ làvrí nɛ̀s̰ɛ ̰̀ near Lavri �gbá làvrí gbà below Lavri ézúvi ̰̀ làvrí zùvi ̰̀ above Lavri We conclude that the postpositional phrases clearly add up to the prosodic domains of the Tone Lowering Rule. 4.3.5. Explaining some exceptions: èfluúɓá and énàg̰á In the above sections a wide range of empirical motivations for the Tone Lowering process have been provided. There exist, however, some problematic cases in which the Tone Lowering extends over two following High tones as in (13) while it is limited to only one as in (14). (Remember at this place that so far disyllabic words lower only the first syllable to the left as in (9)). This case is well transcribed in Hérault’s paper. Consider the following patterns: 134 Legon Journal of the HUMANITIES Volume 20 (2009) (13) High-High tones Possessive construction a. èflúɓá làvrí flùɓà) (Hérault 269) Lavri’s book b. ɛb̀wáb̰á làvrí bwàb̰à Lavri’s finger c. ɛʒ̀�ɓà làvrí ʒɛɓ̀à lavri’s ear d. èŋ��̰ɓá làvrí ŋʊ̀ɔ̀ɓ̰à Lavri’s eye e. l�ɓá làvrí lɛɓ̀à Lavri’s eye brow f. ɛẁl�ɓá làvrí wlɔɓ̀à Lavri’s egg g. ɛs̀ómá̰ làvrí sòmà̰ Lavri’s sickness h. àfruṵ́mṵ́ làvrí frṵ̀mṵ̀ lavri’s donkey i. áj�l� làvrí jɔl̀ɛ ̀ Lavri’s story (14) a. kéwé làvrí kèwé lavri’s crab b. l�w� làvrí lɔẁ� Lavri’s piece of wood c. tává làvrí tàvá Lavri’s bat The examples (13) and (14) contrast because in the first set all the High tones of the second item are lowered, while in the second set only the first tone of the syllable is lowered. Since there is no apparent explanation, I shall explore some possible analyses and provide independent evidence for one type of solution. The cases we are going to deal with are the words ‘èflúɓá’ ‘book’ and ‘èŋ��̰ɓá’ ‘eye’.We explain the contrast by assuming that ɓá is a dimunitive prefix that has lexicalized, though there are no synchronic cases where these words can be used without that prefix. One possible additional explanation is that the consonant ‘ɓ’ is a transparent consonant as has been documented in most lagoon languages, as opposed to depressor consonant. An additional motivation for this hypothesis is the impossibility of analyzing the lowered third syllable of the word èflúɓá as a suffix because it is a borrowing from other Kwa languages where the same word shows up as fluwa or fluwa in most languages (cf. Ahoua 2007 for discussion of the sources of alternating b and w in Kwa). In ɛ̀sómá̰ the same explanation holds, provided we agree with Hérault (1982) that the implosive ɓ is realized [m] before a nasal vowel. Thus, ɛs̀ómá ̰and àfrṵ́mṵ̰́ are to be interpreted phonologically as /ɛs̀óɓá̰/ and /àfrṵ́ɓṵ /. In (14) our class of approximants are opaque to Low Tone spreading. In the examples (15) Tone Lowering spreads on all the disyllabic words although there are no apparent synchronic suffixes: Firmin Ahoua 135 (15) a. ɛs̀�̰b�̰ làvrí sɔ̀b̰ɛ ̰̀ Lavri’s river b. �ɲáŋ̰á̰ làvrí ɲàŋ̰à̰ Lavri’s snake c. ɛŋ̀�̰ɓá làvrí ŋɔ̀ɓ̰à Lavri’s eye d. ɛk̀áwrú làvrí kàwrù Lavri’s sweat e. kpùkpùɲàɲ̰à̰ làvrí kpùkpùɲàɲ̰à̰ Lavri’s fish (sp.) f. àcùfṵ̀à̰ làvrí cùfṵ̀à̰ Lavri’s nail g. adìrà làvrí dìrà Lavri’s bean One can further observe that Low tone spreads across the consonant ‘b’ as the other consonants ‘w’, ’j’and ‘l’ that were grouped together under the same class with the feature ‘sonorant’ by Hérault (1982:266). The latter group is motivated by the nasalization rule. Nonetheless it is still unclear why the examples (15) b. to f. are subject to Tone Lowering unless we hypothesize different diachronic sources for these words1. One further question that may arise at this junction is whether it makes sense to postulate that a Low tone is inserted by morphology to mark the compounding, and then to stipulate that the tone is displaced to the right-standing syllable by a phonological constraint. As I´ve argued here, tonal spreading beyong a transparent consonant is a phonological rule. This leads us to admit that the morphology is governed by morphological conditions, and the phonology by phonological constraints. Such a structure is similar to the type of model that is exemplified in Kiparsky´s (1982) and Mohanan´s (1986) versions of the theory of Lexical phonology. 5. Across-the-board Tonal Spreading: Recursive prosodic words èflúɓá is particularly revealing, because it shows that a Low tone can spread beyond more than a single syllable. What is apparently a significant observation is that there is a transparency of the medial consonant that doesn’t block the spreading. This may suggest that Avikam interprets some types of sonorants as non opaque. However, it’s hard to see why the spreading rule would affect also those words that use all the other regular types of obstruents as in (16) : (16) làvrí [cù] Lavri’s sea làvrí [cù][sɛ̀]̰ Lavri’s sea fish làvrí [cù][ sɛ̀]̰[sɔ̀]̰ Lavri’s home sea fish làvrí [cù][ sɛ̀]̰[sɔ̀]̰[và] Sea fish from Lavri’s home compound 136 Legon Journal of the HUMANITIES Volume 20 (2009) At first sight, one may hypothesize that the phonological structure of the word could give clues. Note in (16) that if we align a sequence of monosyllabic words, just as in Baule (Leben and Ahoua 1997), the High tones of these syllables would be affected. Note also that the very last examples in (16) are composed of one to four monosyllabic words behaving as one prosodic word, leading to the conclusion that prosodic word formation is recursive. Note also that it is exactly the prosodic word that constitutes the domain for the loss of the initial prefixes. To sum up, the prosodic word hypothesis tested by concatenating more than one monosyllabic word results into an across-the-board lowering as in (16). Typical informal acoustic measurements averaged from 3 utterances as pronounced by one single speaker exhibit the F 0 -values in Hertz in (17). For the sake of clarity, I first present in (17) a. the measurement values of syllables with lexical High tones spoken in isolation. Then in (17)b. I indicate the fundamental frequency (Fo) values of sequences of tones of the same syllables in context that illustrate the lowering effect: (17)a. Text: �s�̰ �s�̰ �vá Lexical Tones: H H H H H H Averaged F 0 113 143 110 133 109 120 (17 ) b. Text: Là vrí cù sɛ̰̀ s�̰ và Lexical Tones: L H H H H H Surface Tones L H L L L L Averaged F 0 98 127 98 97 93 85 As we see in (17) a., every second syllable of each word in isolation has a high frequency. As expected in (17) b., Avikam drops the High tone prefixes [ε] in nominal compounding, and in general in all constructions. The table in (17) b. illustrates the progressive lowering of F 0 -values with underlying High tones occurring after the syllable [vri] . Such a lowering process can extend over a larger number of monosyllables with underlying High tones, provided they consist of prosodic words. A pitch extraction of sentence (17) b. gives the following visual representation : Firmin Ahoua 137 FIGURE 3: Pitch track of Tone Lowering in lavrí cú sɛ̰ ̀ s�̰ và To sum up, we would like to propose that the phonological word described here is larger than just a morphological word, and may correlate with a complete noun phrase, unless it contains a coordination (as in [NP and NP]). In that case, the second NP starts a new phonological phrase. The phonological word has thus recursive properties and may coincide with the phonological phrase, in accordance with the Strict Layer Hypothesis. The phonological word is also the minimal domain for prefix truncation (apheresis), whether this operates cyclically or across-the-board. The next questions that arise are now: What is the status of the phonological phrase? How does it differ from the phonological word? 6. The Phonological Phrase The phonological phrase is attested in all verbal phrases that include one object or two objects (arguments). The motivation for this assumptionn is that no Tone Lowering is applied after the verb. 138 Legon Journal of the HUMANITIES Volume 20 (2009) 6.1. Verb Complements constructions Let us illustrate the verb-complements constructions with NP arguments expressing objects, manner, time or location, the High tones of which don´t lower, as we can see in (18) (18) a. éwú lé wr� èdómrí ̰né → éwú lé wr� ̀dómrí ̰né bone-the-has burned -night -the the bone has burned during the night b. áz�̰ sàjé ɛt̀j�wú lé → áz�̰ sàjé ̀tj�wú lé Azoh-has thrown–dog-bone -the Azoh has thrown the dog´s bone c. ɛ ̀sra̰ ̀�s�̰ l� ɛd̀á̰ → ɛ ̀sra̰ ̀`s�̰ n�̰ ˋdá̰ He-has asked for- house-the -price he has asked for the price of the house As we observe in (18) the tones of èdómrí ̰né, ɛt̀j� wú lé, l� dá ̰ are unaccessible to Tone Lowering, which suggests the existence of a prosodic phrase boundary that is opaque to Tone Lowering. An additional evidence of the prosodic phrase is provided by the double object constructions that nicely contrast with possessive constructions. This is discussed in detail in the next section. 6.2. The double object vs. possessive constructions Double object constructions contrast with possessive constructions in the way they interact with Tone Lowering. The beginning of a phonological phrase preserves the prefix of the noun if that noun is in initial position. Consider the contrast in (19) and (20) in which every pairwise example illustrates the blockage of the tone lowering process on the rightmost noun: (19) a. m̀ ná ̰làvrí s�̰ I gave a fish to Lavri b. m̀ ná ̰làvrí s� ̰ I gave Lavri’s fish c. m̀ ná ̰làvrí sáká I gave Lavri rice d. m̀ nà ̰làvrí sáká I gave Lavri’s rice Firmin Ahoua 139 e. m̀ ná ̰làvrí frúɓá I gave Lavri a book f. m̀ ná ̰làvrí frùɓà I gave Lavri’s book (20) a. m̀ b� làvrí ŋɔɓ̀à I hurt Lavri’s eye b. m̀ b� làvrí ŋ�ɓá I hurt Lavri at his eye c. m̀ b� làvrí ɛŋ̀�ɓá I hurt Lavri at his eye The case of blockage to Tone Lowering (that also applies even if there is a monosyllabic clitic pronoun as a second object) can be explained by the syntactic environment. Similar contexts have been shown to operate in Bia languages (cf. Leben and Ahoua 1997). Another strong evidence for the phonological phrase is the optional occurrence of the prefixes in the double object constructions. The drop of the prefixes is obligatory in the possessive constructions, but is obligatorily maintained at the beginning of a prosodic phrase. Compare the examples in (21) (21) m̀ ná ̰làvrí ɛs̀ɛ̀ ̰̰ I gave a fish to Lavri *m̀ ná ̰làvrí ɛs̀ɛ̀ ̰ I gave Lavri’s fish m̀ ná ̰làvri ɛf̀rúɓá I gave Lavri a book *m̀ ná ̰làvrí ɛf̀rùɓà I gave Lavri’s book The preceding examples clearly illustrate the role of the phonological phrase as opposed to the prosodic word, that are both independently required by the rule of prefix drop (apheresis) and the Tone Lowering rules. In the next sections, we would argue that the syntactic constituents are not isomorphic with the phonological word or the phonological phrase. 6.3. The noun and adjective vs. noun and numeral constructions By contrast, adjectives are never lowered after a noun, while numerals are affected by Tone lowering. Note in the examples (22) that once a new phrase is constructed, there is no long distance effect of Tone Lowering on numerals, suggesting that adjectives constitute a new prosodic phrase. (22) ámág̰ó vúcú áz̰á̰ three white mangos ámág̰ó àz̰à ̰vúcú three white mangos ámág̰ó kr�� áz̰á̰ three unripe mangos ámág̰ó a̰z̀à ̰kr�� three unripe mangos 140 Legon Journal of the HUMANITIES Volume 20 (2009) Whenever adjectives are reduplicated to function as intensifiers, no Tone Lowering applies: (23) ámág̰ó vúcú vúcú very white mangos ámág̰ó kr�� kr�� very unripe mangos ámág̰ó ébríákré ébríákr� very red mangos These examples illustrate the recurrent property of the phonological phrases. Let us consider other constructions where the same process occurs, providing independent evidence for the phonological phrase. 6.4. The noun and definite vs. the noun and numeral constructions Avikam has a class of demonstratives, determiners and quantifiers bearing a High tone that does not lower: nínín ‘this’, ébúlé ‘all’, le ‘that’. Notice in (26) that the word ´ébúlé does not lose its prefix, suggesting that definites are part of a new prosodic phrase, similarly to adjectives. (24) �vá l� that compound (we talked about) �s�̰ l� that fish (we talked about) sáká l� that rice éwú lé that bone (25) sáká níní this rice àcúfúá ̰níní this nail lówú níní this corpse (26) ámág̰ó ébúlé all the mangos sáká ébúlé all the rice �d�̰ ébúlé all the villages �s�má ̰ébúlé all the diseases Firmin Ahoua 141 Note that contrary to definite markers, the High tones of numerals are lowered as already illustrated in (22) and repeated in (27) : (27) Citation form In context Gloss áz̰á̰ ámág̰ó àz̰à̰ three mangos sáká àz̰à̰ three pieces of rice �s�má ̰àz̰à̰ three diseases In short, we suggest that the structures of the prosodic word and the prosodic phrase are not directly related to syntactic branching (sisterhood), depending presumably on how we analyse the immediate adjacency position of adjectives (as opposed to numerals and definite markers) to a noun. Apparently the prosodic structure imposes some types of phrasing algorithm that we need to account for in the analysis of Avikam. In the next section we would like to show how the tone lowering constitutes a more complex phonological domain that cannot be captured by morphosyntactic criteria in a straightforward way. We also deduce the relevance of the moraic unit as opposed to the syllabic unit. Conjonction and noun or adverbial constructions1.5 In general, nouns following other nouns or adverbs followed by adverbs with a conjunction constitute individually new clauses and new prosodic phrases. Consider the following examples: (28) a. ègú má̰ ám�̰ → ègú máá̰m�̰ ‘chimpanzee’ Conj ‘me’ ‘a chimpanzee and me’ b. àm�̰ má égù → ám�̰ mégù ‘me’ Conj ‘chimpanzee’ ‘me and a chimpanzé’ c. ɛt̀j� má áƔ� → ɛt̀j� máƔ� ‘dog’ Conj ‘you’ ‘a dog and you’ d. áƔ� má ɛt̀j� → áƔ� mɛt̀j� ‘you’ Conj ‘dog’ ‘you and a dog’ (29) a. lávr�̰ má̰ áz�̰ → lávr�̰ máá̰z̰�̰ ‘there’ Conj ‘here’ ‘ over there and here’ 142 Legon Journal of the HUMANITIES Volume 20 (2009) b. émrílí má ɛs̀ín̰�̰ → émrílí mɛs̀ín̰�̰ ‘today’ Conj ‘yesterday’ ‘today and yesterday’ c. ézúvi ̰̀ má �sám → ézúvi ̀ ̰m�sám ‘above’ Conj ‘below’ ‘above and below’ d. ávázù má �nám → ávázù m�nám ‘before’ Conj ‘behind’ ‘ahead and behind’ e. ɛf̀r�̰ má ɛd̀já → ɛf̀r�̰ mɛd̂já ‘left’ Conj ‘right’ ‘left and right’ As we see, the prosodic pattern is [NP [ma NP]]. It is also crucial to note again the similarities with most Tano languages such as Anyi, Nzema and Akan. 7. Tonal foot as a constraint on Tone Lowering and the account of CVV vs. CGV words Avikam has a class of words that may be interpreted either as CVV or CGV. Hérault (1982) has chosen to transcribe them as CGV. The challenging question related to these syllables is to explain why complex syllables with High tones occurring as the second of a disyllabic word are realized with a rising tone (Low- High) as in (30) when we expect an across-the-board tone lowering: (30) d�ví� làvrí d�ví� l� Lavri’s evening lézíé làvrí lèzìé lé Lavri’s tail Note that by contrast the complex syllables CVV carrying High tones are entirely lowered in (31) : (31) ébíé làvrí bìè Lavri’s pot �b�r� làvrí bʊ̀rɛ̀ Lavri’s heart �díá làvrí dìà Lavri’s tooth stick ás�á làvrí sʊ̀à Lavri’s mouse Observe first of all that the significant similarity between (30) and (35) is the fact that the prefix drops and that (30) and (31) are disyllabic, though in (30) the Low tone doesn´t spread to the last High tone as we can read off the right-hand Firmin Ahoua 143 columns. The proposed explanation is that two tones represent a tonal foot that constitute a domain of Tone Lowering. So, assuming that the complex syllable is to be interpreted as a CGV syllable, we would incorrectly predict a Low tone rather than a rising tone in (30) because only vowels are tone bearing units. However, as just argued, if we choose to interpret the complex syllables as a sequence of moras CVV, it becomes straightforward to assume that each mora can carry its own tone, and therefore that each tone can undergo its own process. Support for this analysis is provided by words such as those in (32) where sequence vowels co-exist with the adjunction of a nominal suffix marker �̰: (32) k� ‘to fight’ kʊ̰̀ɛ ̀ ̰ vs * kw̰ɛ̀ ̰battle’ ác�̰‘to dart’ ác��̰ vs * kwɛ̀‘̰dart’ Avikam verbs that have the structure CVV or CCV bear Low- High tones while monosyllabic verbs carry unambiguously only High tones, and monosyllabic words generally do not carry contour tones. To my view, this points to the disyllabic nature of these words. A similar fact has been widely attested in most Kwa languages, especially with verbs in isolation and in their morphophonological behavior at the tonal and segmental levels (cf. Ahoua and Leben 1999). So, assuming that CVV and CCV words are to be analyzed as disyllabic raises the question of how to represent the fact that these syllable types attract Tone Lowering. One natural candidate solution is to suggest that the CVV are phonologically to be interpreted as disyllabic words just as CVCV words and that their tones constitutes a tonal foot, a term independently motivated in some other African languages (cf. Leben 1997). If this analysis is correct, it is possible to generalize and to propose a Tonal Foot Formation algorithm as in (33): (33) Tonal foot formation Every High tone of a disyllabic word is a tonal foot.a. Tone Lowering applies on the first two High tones.b. As we see, this domain is purely phonological. Notice however that the just formulated Tonal foot contrasts with the disyllabic prosodic words in (13) and (14) above which we repeat below. (34) a. kéwé làvrí kèwé lavri’s crab 144 Legon Journal of the HUMANITIES Volume 20 (2009) b. tává làvrí tàvá Lavri’s bat c. l�w� làvrí lɔẁ� Lavri’s piece of wood d. l�ɓá làvrí lɛɓ̀á Lavri’s eye brow Let´s note that patterns such as (14) apply in High tones of adjectives adjoined to nouns as qualifiers (ebhipa “small”, eson bhìpa “small river”). If we analyze (30), (31) as CVV, the solution is straightforward because we have independently motivated tonal foot in Avikam that behave the same way, allowing to account for a simpler generalization. To sum up, the present section has provided empirical evidence for the significant role that the Lowering plays in the phonology of the language. Assuming the Strict Layer Hypothesis, the tonal foot is projected below the prosodic word and therefore can be licensed to undergo partial Tone Lowering. The empiral consequences for this approach is the emergence of the constraint that disyllabic words are maximal prosodic domains that contrast with monosyllabic words. 8. Leftward docking and further evidence of (segmental) phonological conditioning This section considers segmental influence on tonal rules and attempts to account for the data collected by Herault (1982). The reason for this section is to show that motivation for floating tones as a morphosyntatic morpheme are rather weak in other domains, and that a domain construction or domain alignment might be a necessary algorithm to account for the facts in Avikam. The prosodic domains in Avikam cannot be always directly read off the syntactic constituents, and do not rely on a morphosyntactic trace such as the floating tone generally documented in Niger-Congo languages. Notice in the examples (35) that the Tonal lowering on the right is not applied. (35) m̀ ɲɔ̀ ̰àlábá I caught a pearl m̀ ɲ�̰ ásúá I caught a mouse On the contrary, tones on the right tend to dock on the syllables on the left, contradicting the directionality of the docking hypothesis found in possessive constructions. This corresponds to Hérault’s (1982 :273) “Alpha Rule” that reads : Firmin Ahoua 145 “la dernière (ou unique) syllabe du verbe adopte le ton inverse de celui qui lui fait suite si elle n’en est séparée par une consonne( ou plusieurs). “ According to this rule, the last syllable of the verb copies the tone of the following item unless that item has one or more consonants that block the copy process. The copying rule is exemplified by the following examples from Hérault (1982:273). (36) Underlying Surface � ɲ�̰ ɛs̀�̰ � ɲ�̰ ̀s�̰ Let him drink water! lɛ ̀ɛk̀á lɛ ̀ ̀ká go (somewhere) for a trip! The examples (36) are injunctive constructions. The subject pronoun and the verbs are assigned a High tone and floating tones are not allowed to dock neither to the right nor to the left (Hérault 1982:275). Notice also that the floating Low tone downsteps the High tone of the object nouns, and this is explicitly transcribed by Hérault (1982:275) On the other hand, Avikam shows other rules sensitive to segmental quality, the effect of which is to trigger tonal polarity after apocope. If the object noun has an initial prefix vowel, the tone of that vowel copies onto the lefthand verb. However, if an initial vowel is not available, and that the object noun starts with a consonant, that consonant triggers the change of the tone of the verb into an opposite level. I refer to this total change of level as tonal polarity. Such a rule has been found to apply in other Kwa languages such as Ega, in the imperative mode. (37) m̀ ɲ�̰ grìgbé I caught a grass cutter m̀ ɲɔ̀ ̰dábʊ̀ I caught a duck m̀ ɲ�̰ ásúá I caught a mouse m̀ ɲɔ̀ ̰àlábá I caught a pearl When we add to this evidence the fact that downstepped High exists morpheme internally, and that there again the floating low tone does not dock, the morphological solution of an associative marker may seem ad hoc. 9. Conclusion We have attempted to argue that Tone Lowering in Avikam plays a significant role in understanding the prosody of Avikam, because it helps to identify the tonal foot (true bisyllabic words), the recursivity of prosodic words (sequences of 146 Legon Journal of the HUMANITIES Volume 20 (2009) monosyllabic words) and the existence of the prosodic phrase (adjective, numeral consitutents) and a disyllabic word constraint. The present results, while they clearly provide evidence for a hierarchical prosodic structure, do diverge to some extent from the strong version of Selkirk´s (1990:180) Strict Layer Hypothesis. According to this model every category must be immediately included into a higher category. We argued however that recursivity should be licenced in Avikam, similarly to Baule (Leben and Ahoua 1997). At the tonal level, the tonal foot made up of two High tones in a non-prefixed disyllabic word undergoes Tone Lowering of the first High tone. The prosodic word composed of a prefixed monosyllabic word lowers all sequences of High tones across-the-board and iteratively. The phonological phrase is opaque to Tone Lowering while the phonological word is transparent to that rule. Phonological phrases generally correlate with clauses of noun phrases (whereby every coordination conjunction starts a new phonological phrase), determiner phrases, adjective phrases, prepositional phrases. In verbal constructions, all verbal complements (adverbs, objects and double objects) are phonological phrases, a phenomenon that is quite common in Kwa languages, especially in Tano. Phonologically Tone Lowering makes it possible to distinguish morphosyntactic domains from phonological domains. Phonologically Tone Lowering helps to distinguish between a CjV (consonant palatal glide and vowel) and a CiV (consonant vowel vowel) analysis of the syllable structure. The evidence has been shown to be widely attested in a great number of Kwa languages (cf. Ahoua and Leben 1999). The phonology of Avikam shows the conspiracies between segments and tones that make the language comparable to Adiukru, a geographically and genetically closely related language. Avikam also presents a rare case of tonal inversion triggered by voiced consonants and suggests floating tones in the verbal constructions. From a comparative and typological perspective, we can suggest that Avikam and Baule, both Kwa languages, have in common that they tend to distinguish between monosyllabic word and polysyllabic ones, and most crucially that the former are generally incorporated into recursive prosodic words. The cases considered here point to the relevance of prosody in modelling the intonation and the tonal system of a language. I conclude by pointing out that for literacy purposes most of the present results can be easily incorporated into a proposal for a straightforward orthography, for instance in the notation of compound nouns and the default tones. Noun compounds should be marked with hyphens. Finally since Low tones are triggers for tonal changes, they should be written, whereas High tones being default tones should be left unspecified, in accordance with Rongier (2002). Firmin Ahoua 147 Acknowledgement I would like to thank the Alexander von Humboldt Foundation for support. I am also grateful to my colleagues, Late Gbery Aimé, Jacques Rongier and my students, especially Armand Niaba who has kindly made his parents and relatives available for my long questionnaires. I have also benefited from invaluable comments from Will Leben. Needless to say, I am alone responsible for flaws and misinterpretations contained in the paper. 148 Legon Journal of the HUMANITIES Volume 20 (2009) NOTES According to our informant, Diecket Moise, 1. �ɲáŋ̰á̰”snake” may be a compound of �ɲá̰“animal” ŋá̰ “looking for quarrels” and ɛ̀s�̰b�̰ as a lexicalized compound of ɛs̀�̰ “ water “ and b�̰ “sweet” (meaning sweet water or river). ɛk̀áwrú “sweat” may be composed of ɛ ̀ká”time” and wrú “heat”. Firmin Ahoua 149 REFERENCES Ahoua, F. 2007. Reconstruction of Consonants in Bia languages . Innovation or Retention, in M. E. Kropp Dakubu, G. Akanlig-Pare, E.K. Osam and K. K. Saah, Studies in the languages of the Volta Basin. Vol 4,, part. 2. Legon : Combert Impressions. Ahoua, F. 1986. The Autosegmental Representation of Tones in Akan : More Evidence for the Tone Mapping Rule with Reference to Baule. In K. Bogers, H. van der Hulst and M. Mous (eds), The Phonological Representation of Suprasegmentals. Dordrecht : Foris. Ahoua, F. and Leben, W.1999. Syllable Structure in Western Kwa. Paper presented at the 21st Conference of the West African Linguistics Society, Abidjan, June 1999. Ahoua, F. and Leben, W. 2007. (eds). Morphophonologie des langues kwa de Côte d´Ivoire. Cologne : Rüdiger Köppe. Akinlabi, A. and Urua, E-A. 2003. Foot structure in the Ibibio Verb. In Journal of African Languages and Linguistics. 24/2, pp. 119-160. Dolphyne, F. 1986. Tone and Grammar in Akan. The tone of possessive Construction in the Asante dialect. In K. Bogers, H. van der Hulst and M. Mous (eds), The Phonological Representation of Suprasegmentals. Dordrecht : Foris Duponchel, L. 1971. L’avikam. In G. Dumestre (ed.) Atlas Linguistique de Côte d’Ivoire : les langues des régions lagunaires . Abidjan: ILA-ACCT. , pp. 255-276. Hérault, G. 1982. L’Avikam. In: Atlas des Langues Kwa de Côte d’Ivoire ed. By G. Hérault. Abidjan: ILA-ACCT. 255-276. Inkelas, S. and Draga, Z. 1990. The Phonology-Syntax Connection. Chicago and London: University of Chicago Press. Kiparsky, P. 1982. Lexical Morphology and Phonology. in Yang, I. s. (ed). Linguistics in the Morning Calm. Seoul : Hanshin. 150 Legon Journal of the HUMANITIES Volume 20 (2009) Ladefoged, P. 1968. A Phonetic Study of West African Languages. London : Cambridge University Press. Leben, W. 1996. Tonal Feet and the Adaptation of English Borrowings into Hausa. Studies in African Linguistics 25, Nr 2. 1, pp. 39-154. Leben, W. and Ahoua, F.. 1997. Prosodic Domains in Baule. Phonology 14. Number 1, pp. 113-132. Kadio-Morokro, C. 1978. Phonologie de l’avikam.. Abidjan : Mémoire de Maîtrise. Mohanan, K. 1986. The Theory of Lexical Phonology. Dordrecht : Reidel. Nespor, M. and Vogel, I. 1986. Prosodic Phonology. Dordrecht: Foris. Selkirk, E. 1978. “On prosodic structure and its relationship to syntactic structures”. Nordic Prosody 2, pp., 111-140. Selkirk, E. 1990.”On the Nature of Prosodic Constituency”, in J. Kingston and M. E. Beckman (eds.) Papers in Laboratory Phonology, vol. 1. Cambridge: Cambridge University Press. pp. 179- 200. Truckenbrodt, H. 1999. “On the relation between syntactic phrases and phonological phrases”..Linguistic inquiry 30, pp. 219-255. Rongier, J. 2002. Parlons Avikam. Paris: Harmattan. Urua, Essien E. A.. 2007. Ibibio Phonetics and Phonology. Port Hartcourt: Grand Orbit Welmers, W. E.1963. “Associative a and ka in Niger-Congo”. Language 39, pp. 432-47. Williamson, K.. 1986. The Igbo Associative and Specific Constructions. In K. Boegers, H. van der Hulst, and M. Mous (eds.), the Phonological Representation of Suprasegmentals. Dordrecht: Foris. 151 DOUBLE PLURALITY IN CODESWITCHING Evershed Kwasi Amuzu1 Abstract Introduction1. 1 Dr. Evershed Amuzu is a Lecturer in the Department of Linguistics, University of Ghana. This paper deals with the phenomenon of double plurality in codeswitching, with illustrations from Ewe-English Codeswitching (CS). It shows that English nouns (but never Ewe ones) may take two plurals, the English -s and its Ewe counterpart wó. –s always occurs on the stem of the noun while wó occurs either immediately after –s or a few slots away. The paper demonstrates that the English noun-and- plural units are consistently embedded in Ewe-based NPs in which Ewe modifiers of the English nouns occur in slots associated with them in monolingual Ewe NPs. While –s may be dropped from mixed NPs that already show double plurality, the dropping of wó from such NPs makes them unacceptable. Three theoretical questions are asked in our quest to explain this plural doubling phenomenon. One is why it involves only English noun heads. The second relates to why the two forms emerge as plurals even though –s is redundant. The third one is about the nature of language production involved: what bilingual processes are involved in the phenomenon of double plurality? It is shown that the two plurals arrive in their respective positions in the mixed NPs via separate paths in language production. The distribution of –s relates to processes that are conceptual (i.e. semantic-pragmatic) in nature. On the other hand, the distribution of wó relates to processes that are morpho-syntactic in nature. Another issue briefly explored is what this bilingual phenomenon reveals about the linguistic properties of the plural category in monolingual NPs. The paper ends with a discussion of the role that language typology plays in determining whether plurals may be doubled or not doubled in CS. Double morphology is the codeswitching (CS) phenomenon in which a grammatical category is expressed twice with equivalent grammatical elements from two languages (see Appendix 1 for examples from Ewe-English CS). Double plurality is the type of double morphology by which a nominal from one language is marked with two plural morphemes, one from the language 152 Legon Journal of the HUMANITIES Volume 20 (2009) of the nominal and the other from another language. The paper explores this phenomenon with illustrations from Ewe-English CS. Scrutiny of data shows that only English nouns are used in mixed NPs showing double plurality; English nouns may take two post-modifying plurals: the English plural –s, which is always bound to the noun stem, and the Ewe plural wó, which may follow –s immediately (as in examples 1 and 2 below) or be separated from it by other post-modifiers (as in examples 3 and 4).1 (1) Woawo-ɛ nye-na [{executive -1 member 0 -s +1 } wó +5 ] kple [{patron 0 -s +1 } - wó +5 ] 3PL -FOC be-HAB PL PL and PL PL ‘They normally are the executive members and patrons.’ (Asilevi 1990: 70) (2) [Nye -2 {younger -1 brother 0 -s +1 } -wó +5 kata) +7 ] wó shave- na… 1sg PL PL all 3pl HAB ‘All my younger brothers, they shave…’ (Amuzu 1998:72) (3) Headmaster la inform student -aɖe -wó be wó-a- label DEF INDEF-PL COMP 3PL-SUBJ [{textbook 0 s +1 } yɛyɛ +1 -a +4 -wó +5 ]. PL new the PL ‘The Headmaster informed some of the students to label the new textbooks.’ (Asilevi 1990: 34) The rest of the paper is in three sections, which deal with patterns of double plurality, explanation of the phenomenon, and the significance of the study respectively. Patterns2. Evershed Amuzu 153 (4) Nukata-ɛ me iron [nye -2 {shirt 0 -s +1 } eve +2 sia +4 wó +5 hã +7 ] o? Why FOC 2sg.NEG 1sg PL two this PL too NEG ‘Why didn’t you iron these two shirts of mine as well?’ In example (1), the English plural –s and the Ewe counterpart -wó occur in succession after both member and patron. The same pattern emerges in (2), where the two plurals follow brother. In examples (3) and (4), however, the plurals are separated by other morphemes. In (3), two Ewe morphemes (the adjective yɛyɛ ‘new’ and the definite article –a ‘the’) separate –s and wó; and in (4), they are separated by the Ewe cardinal eve ‘two’ and the proximal demonstrative sia ‘this’. Carefully examined, three critical observations may be made about these CS patterns: Observation 1: There are two brackets defining the structure of each mixed NP. The inner bracket contains an NP constituent that is well-formed in English: the English morphemes occur in accordance with word order in the English NP, where the head noun (in 0 slot) is immediately followed by the plural marker (in +1 slot). Where an English adjective is involved, as in example (1) above, it appears in its ‘home’ adjective –1 slot. This English constituent is embedded within a larger Ewe-based NP structure, which is represented by the outer bracket. This larger NP is Ewe-based because the order of morphemes / constituents is consistent with what obtains in the monolingual Ewe NP, which is captured in Appendix 2. It should be noted in example (4) above, for instance, that nye ‘my’ occurs in its pre-modifier -1 slot and is followed by the English NP unit in the 0 slot of the Ewe noun head. This unit is then followed by various Ewe post-modifiers in their respective ‘home’ slots: the cardinal eve ‘two’ in +2 slot, the demonstrative sia ‘this’ in +4, the plural wó in +5, and the intensifier hã ‘too’ in +7. This word order is significant, and we shall be returning to it later in the paper. 154 Legon Journal of the HUMANITIES Volume 20 (2009) Observation 2: The second observation about the CS patterns presented above relates to what happens when either plural is dropped. While the English plural marker may be dropped and the structure will remain acceptable, the omission of the Ewe counterpart would render the sentence unacceptable. This is shown in the following versions of examples (1) – (4). The omission of either –s or wó is signaled with the symbol Ø. Note that while the first of each pair of examples is acceptable when only wó is present, the second of each of them is unacceptable because only –s is present. Versions of (1): (5) a: [{executive -1 member 0 Ø +1 } wó +5 ] kple [{patron 0 Ø +1 } - wó +5 ] -- PL and -- PL b: [ {executive -1 member 0 -s +1 } *Ø +5 ] kple [{patron 0 -s +1 } *Ø +5 ] PL -- and PL -- ‘executive members and patrons.’ Versions of (2): (6) a: [Nye -2 {younger -1 brother 0 Ø +1 } - wó +5 kata) +7 ]... 1sg -- PL all b: [Nye -2 {younger -1 brother 0 -s +1 } *Ø +5 kata) +7 ]... 1sg PL -- all ‘All my younger brothers…’ Versions of (3): (7) a: [{textbook 0 Ø +1 } yɛyɛ +1 -a +4 - wó +5 ] -- new -the PL b: [{textbook 0 -s +1 } yɛyɛ +1 -a +4 * Ø +5 ] Evershed Amuzu 155 PL new -the -- ‘the new textbooks’ Versions of (4): (8) a: [nye -2 {shirt 0 Ø +1 } eve +2 sia +4 wó +5 hã +7 ] 1sg -- two this PL too b: [nye -2 {shirt 0 -s +1 } eve +2 sia +4 * Ø +5 hã +7 ] 1sg PL two this -- too ‘these two shirts of mine too’ The acceptability of the (a) examples as opposed to the (b) examples means that only wó is a critical, i.e. grammatically obligatory, marker of plurality in the mixed NPs. In other words, going back to examples (1) to (4) we may claim that –s is redundant from the point of view of grammar even though it is the one bound to the English noun stem. Observation 3: The third observation is that the double plurality cannot involve an Ewe noun head. For instance, examples (9) and (10) below, versions of (3) and (4) respectively, are unacceptable because the original English noun heads have been replaced with their Ewe counterparts: (9) a: ...be wó-a- label agbalẽ0 *-s+1 yɛyɛ+1-a+4 -wó+5. COMP 3PL-SUBJ PL new -the -PL ‘… that they should label the new textbooks.’ b: ...be wó-a- label agbalẽ0 yɛyɛ+1-a+4 -wó+5. (10) a: Nukata-ɛ me iron [nye -2 awu 0 *-s +1 eve +2 sia +4 wó +5 hã +7 o? Why FOC 2sg.NEG 1sg two this PL too NEG 156 Legon Journal of the HUMANITIES Volume 20 (2009) ‘Why didn’t you iron these two shirts of mine as well?’ b: Nukata-ɛ me iron [nye -2 awu 0 eve +2 sia +4 wó +5 hã +7 o? In the light of these observations, we need to answer three theoretical questions. First, why does the doubling of plurals happen only when an English noun head is involved? Second, why are the two plural forms needed on an English noun head even though only wó, as noted, is the critical marker of plurality? Third, what bilingual processes can be assumed to be involved in double plurality? That is, how do the two plural forms come to be repeated on one head noun? The Double Plurality3. 3.1 Theoretical assumptions The answers to these questions derive from the fact that Ewe-English CS is composite CS; that is, the languages involved play complementary abstract grammatical roles in the determination of the patterns in which morphemes from both are distributed (Amuzu 2005a, 2005b, and forthcoming). Underpinning the notion of composite CS are four theoretical assumptions: 3.1.1 Assumption 1 Our first assumption is what Myers-Scotton’s 4-M model stipulates: that there are four types of morphemes in human languages (see Myers-Scotton and Jake 2000: 3ff and Myers-Scotton 2002: 72ff). These are: content morphemes: a. nouns, verbs, adjectives, adverbs, and a few others. early system morphemesb. : grammatical elements that have conceptual affinity with their content morpheme heads, e.g. verb satelites (e.g. INTO in LOOK INTO meaning ‘to consider’), the pluralizer of nouns, demonstratives, intensifiers, etc Evershed Amuzu 157 late bridge system morphemesc. : elements that provide grammatical links between two units, e.g. copulas and possessive linkers late outsider system morphemesd. : critical grammatical elements, e.g. tense, modal, and aspect (TMA) markers, agreement inflections, case markers, etc 3.1.2 Assumption 2 The second assumption is that CS constituents are better understood in terms of the nature of their underlying abstract grammatical structures. This assumption stems from the view that the basis of syntax is the abstract representations underlying lexical items known as lemmas (Myers-Scotton and Jake 1995, 2001 and Myers- Scotton 2002). Briefly put, a lemma is the non-phonological set of information about a lexical item in a language which informs the lexical item’s distribution as a surface-level element. Lemmas are supposed to be stored in speakers’ mental lexicon of a language. They are thus language-specific. According to Myers- Scotton, “Lemmas contain lexical rules and these rules contain all the necessary information to realize surface constructions” (2002: 14). Specifically, a lemma consists of three subsets of lexical rules concerning the lexical item’s meaning and distribution: lexical-conceptual structure• , i.e. details about the lexeme’s semantic and pragmatic properties (e.g. does a noun encode Agent, Patient, or Experiencer?; and does a verb encode Action, State, or Process?) predicate-argument structure• , i.e. details about the lexeme’s syntactic properties (namely details about its thematic structure that would be mapped on to grammatical relations); i.e., for example, whether a noun conceptualized as Patient is to be expressed as Subject or as Object. morphological realization pattern• , i.e. specifications about language- specific devices—like word order restrictions, agreement, tense / aspect marking system, etc—for realizing the lexeme’s grammatical relations with other lexemes in surface configurations, e.g.: Must a Subject come before its verb or may it occur elsewhere?; Are case-markers required on the Subject?; etc 158 Legon Journal of the HUMANITIES Volume 20 (2009) As will be shown shortly, this notion of lexical structure is useful in explaining how surface CS configurations containing double plurals stem from the nature of the lemma information about the lexemes (e.g. the English nouns in the mixed NPs). 3.1.3 Assumption 3 The third assumption is that language production is modular, involving four stages of operation, also called levels. These are the conceptual level, lemma level, functional level, and surface/positional level. In table 1 below, details of what transpires at each stage are outlined in columns on the right. The first stage, the conceptual level, is pre-linguistic. Only universally available concepts, entities, propositions and the like are therefore activated at this stage. The second stage, the lemma level, is the first linguistic level of production. Here, there is a conceptual activation of lemmas underlying language-specific content morphemes selected to encode the entities, predicates, etc that are mapped from the conceptual level. In some cases, the conceptual activation of a content morpheme also triggers the conceptual activation of an early system morpheme. For example, the form of a plural morpheme, which is an early system morpheme, may be selected to go with that of a noun because that plural item points (semantically) to the fact that there are two or more of the entity to which the noun refers. The third stage, the functional level, involves morphosyntactic procedures aimed at projecting slots for the selected content morphemes. The projection of slots for the content morphemes is based on their language-specific lemma information about predicate-argument structure and morphological realization patterns. In other words, the forms of critical grammatical elements, referred to as “late system morphemes” under the 4-M model, are selected to express the content morphemes’ grammatical requirements (i.e. their predicate-argument structures and morphological realization pattern features). Word order issues are also decided here. In fact, what happens at the positional level (the fourth and final stage) is the articulation of grammatical decisions already made at the functional level. Evershed Amuzu 159 Further details of the assumed production process are outlined in table 1 below: Table 1: The language production model (cf. Amuzu’s [2005a:20-21] adaptation of tables from Myers-Scotton and Jake [2001] and Myers-Scotton [2002]) Conceptual Level [Speaker goes into bilingual mode (Grosjean 2001).] Pre-verbal / pre-linguistic speaker-intentions• are made, consciously or unconsciously. Such intentions, which consist of universally • available semantic and pragmatic information about entities, roles, states, processes, degrees, etc., are conflated as specific semantic/pragmatic (SP) feature bundles, which are necessarily language-specific. . Lemma Level [Both Ewe and English content morphemes and early system morphemes may be conceptually activated] Conceptual activation• of content morphemes (e.g. verbs, nouns, adjectives, adverbs, some pre/post-positions, some pronouns, etc). That is, language-specific SP feature bundles activate entries in the mental lexicon called lemmas, which support the realization of the content morphemes. The content-morpheme lemmas may also • point to lemmas supporting early system morphemes - e.g. plurals, intensifiers, definiteness markers, some adpositions, some verb satellites (e.g. up as in ‘to breakup’, etc). That is, early systems morphemes are also conceptually activated to semantically define their content morpheme heads (e.g. a plural specifies that there are more than one of the entity that a noun encodes). Conceptual activation highlights the • lexical-conceptual structure of a content morpheme or the unit that is formed by a content morpheme and an early system morpheme. 160 Legon Journal of the HUMANITIES Volume 20 (2009) Functional Level [Only Ewe morphosyntactic procedures may be deployed to guide the framing of mixed constituents, because of the SMP and the MOP.] Syntactic procedures• take place as information about the predicate-argument structures of the various content morphemes becomes available. The procedures involve the mapping of thematic structure of content morphemes (e.g. Agent vs. Patient) onto grammatical relations (e.g. Subject vs. Object). Morphological procedures• also take place as morphological realization information becomes available. This involves the deployment of language-specific devices for word order, agreement, tense / aspect / mood marking, case marking, negation, etc. Late system morphemes• —or functional elements—are selected to express the required grammatical categories. Positional / Surface Level [Ewe-based mixed constructions are produced.] Phonological and Morphological realizations• take place: i.e. there is articulation of surface structure morphophonemic forms. The crucial point here is that because language production is characterised by these definable stages of operation, two languages in intensive contact have opportunity to interact in specifiable ways, especially at the second and third stages of production, to jointly constrain the structures of surface constituents in which their content morphemes occur. The nature and sequence of such an interaction are captured in the left column in the table above and discussed further in the next subsection. Evershed Amuzu 161 3.1.4 Assumption 4 Our fourth, and final, assumption is that Ewe-English CS is composite CS, the kind of CS in which the two languages share responsibility for building the structure of bilingual constituents (i.e. clauses and phrases). The idea is that there is systematic interaction between the two languages at the lemma and functional levels. This idea is captured in the brackets in the left column in table 1 above. To begin with, the assumption is that for a bilingual constituent to be produced, the speaker must enter what Grosjean (2001) calls “bilingual mode” at the conceptual level (ostensibly because the social setting motivates the interchangeable use of the two languages). At the lemma level, which, as noted elsewhere, is the first linguistic level, both English and Ewe content morphemes are selected because the speaker is already in bilingual mode. This means that the two languages are switched on to direct the activation of their respective content morphemes. The Ewe and English content morphemes are conceptually activated in their own rights just as they would have been were the speaker in monolingual mode. It is language-specific lemma information about each content morpheme that becomes available at this stage. At the functional level, each content morpheme’s language-specific predicate- argument structure and morphological realization pattern features also become salient. However, once all aspects of the language-specific lemma information of Ewe and English content morphemes have become salient, only Ewe remains active at this level. What happens next is in my opinion the most important process in this composite CS: Ewe alone serves as the source of morphosyntactic means for expressing the content morphemes’ lemma requirements. That is, even English content morphemes’ English-origin lemma requirements have to be expressed using fitting Ewe morphosyntactic means. Two principles operate at this level to ensure that Ewe robustly performs this morphosyntactic responsibility. They are the Morpheme Order Principle (MOP) and the System Morpheme Principle (SMP), both of which have been adapted from Myers-Scotton (1993: 82). 162 Legon Journal of the HUMANITIES Volume 20 (2009) With the MOP, the principle is that in any bilingual constituent that includes a singly-occuring content English morpheme and any number of Ewe morphemes, the order of the morphemes will be that of Ewe. Of course, a decision as to where in an Ewe structure to place an English verb, for example, will depend for instance on its valency and morphosyntactic properties in English. For the SMP, the principle is that only Ewe critical grammatical elements may be picked to express structural requirements that must be fulfilled in the distribution of both Ewe and English content morphemes. This means that all late system morphemes required in the distribution of a lone English content morpheme in bilingual context must come from Ewe. As Amuzu (2005a, 2005b, and forthcoming) have demonstrated, because the English content morphemes’ English-origin subcategorization features are factored into any quest for slots for them in mixed constructions, they consistently occur in slots in Ewe-based structures where Ewe morphemes with analogous subcategorization features occur. Those Ewe morphemes they share slots with may or may not be their equivalents; indeed it is not uncommon to find English content morphemes that occur in slots that do not match those in which their Ewe counterparts occur. Despite the dominant role that I perceive Ewe plays at the functional level, I designate Ewe-English CS as composite CS. This is in recognition of the role English plays in providing lemma information about its content morphemes. As noted, all English content morphemes retain their lemma properties in CS contexts. However, claiming Ewe-English CS to be a case of composite CS amounts to a rejection of the notion expressed in Myers-Scotton (1993) that Ewe-English CS is “classic CS”. The point of departure is this: in Ewe-English CS as classic CS, Ewe would serve as the sole source of relevant lemma information that informs the distribution of even singly-occurring English content morphemes. That role would designate Ewe as a one-language matrix language (ML) and make English the embedded language (EL). It is stipulated in Myers-Scotton (1993, 2002) that EL content morphemes may only be fully integrated into ML grammatical frames if lemmas supporting them are sufficiently congruent Evershed Amuzu 163 with those supporting their ML counterparts. In Amuzu (2005a and 2005b), for instance, it is demonstrated that this expectation does not explain all Ewe- English CS data and that the English content morphemes get fully integrated in the Ewe-based frames on the basis of their own lemma information. It is this together with the other assumptions discussed above that inform the explanation of double plurality presented below. 3.2 Explanation First of all, let us recall that the structures of mixed NPs showing double plurality—exemplified in (1) – (4)—are based on the Ewe NP structure, captured in Appendix 2. We have demonstrated that except for English modifiers (e.g. the plural –s and adjectives), which occur in relation to the English noun in occordance with morpheme order in the English NP, all Ewe modifiers occur in their ‘home’ slots from the point of view of the Ewe NP. The English NP unit, however large, is located in the 0 slot associated with the noun heads in the Ewe NP. Example (11) is further illustration. (11) Ale be [eƒe -2 nenem -1 [ {finding 0 -s +1 } ma +4 wó +5 ] ná be wo vá nɔ famous. So that his such PL that PL make COMP 3sg come be ‘So, those specific findings of his made him famous.’ Here, eƒe ‘his’ and nenem ‘such’ occur in their pre-modifier home slots, the -1 and -2 slots respectively. After the English unit in the inner bracket (including the head noun finding, which appears in the 0 slot), the post-modifying Ewe morphemes occur in their traditional slots in the Ewe NP: the demonstrative ma ‘that’ in the +4 slot and the plural wó in the +5 slot. The explanation for this morpheme distribution pattern structure (i.e. in the larger NP) is the morpheme order principle (MOP) outlined in the previous section. The Ewe word / constituent order defines the overall structure of the larger NP. 164 Legon Journal of the HUMANITIES Volume 20 (2009) We would also recall from section 2 that although the two plurals do co- occur with an English noun (see examples 1 to 4) only wó is grammatically vital. As noted, evidence of this fact is that while wó may not be deleted from any of the four examples, -s may be deleted (see examples 5 to 8). This indispensability of wó in NPs showing double plurality is due to the system morpheme principle (SMP), which ensures that only critical Ewe grammatical morphemes are picked to express such grammatical information in CS constructions. What is not explained by this account is why –s, though not grammatically vital, is permitted to also inflect the English noun head. A clue to why –s appears in the NPs lies in the fact that –s accompanies only English noun heads and never Ewe noun heads (see examples 9 and 10 above). This happens because of what transpires at the lemma level. When an English noun is conceptually activated, its underlying lemma triggers the conceptual activation of –s as well, because –s has semantic tie with the noun. The selection of –s on conceptual or semantic ground is, however, not recognized during functional level processes, which are about grammatical procedures for satisfying predicate-argument structure and morphological realization requirements of content morphemes. The fact here is simple: if plurality is to be marked on an English noun selected for CS, then because of the MOP its slot has to be the +5 slot in the Ewe NP. -s does not oocupy this +5 slot, and the SMP ensures that it is filled by the Ewe plural wó. This is why wó is obligatory in (1) – (4) as shown in the versions in (5) – (8). The doubling of the plurals draws attention to their different functions and the different times during language production when their forms get picked. –s is picked at the lemma level because of its conceptual / semantic tie with the English noun head. On the other hand, wó is picked at the functional level to fulfil a structural need for an Ewe form in the plural +5 slot. This explanation is further supported by two other types of mixed NP in which plurality is expressed. One type includes mixed NPs in which irregular English plural nouns are used. Such nouns are lexicalizations of the semantic affinity between the nouns and the plural. They represent the fussion of the nouns and the plural. In Evershed Amuzu 165 Ewe-English CS, these irregular English nouns are used intact although wó too consistently appears in the +5 slot. This is shown in examples (12a) and (13a). Note that (12b) and (13b) are unacceptable because the singular forms of the irregular English nouns are used: (12a) [{Nice -1 children +0 } ya +4 -wó +5 ] a, mia ga dzi ɖe this PL TP 3PL REP give_birth some a-kpe wó o a? FUT-add 3PL NEG Q ‘These nice children, won’t you have some more to add to them?’ (Amuzu 2005a: 221) (12b) Nice [*child ya –wó] a, mia ga dzi ɖe… (12c) Nice [children ya *Ø] a, mia ga dzi ɖe… (13a) Wo be [men +0 wó +5 ] ne nɔ afiyi ne [women +0 wó +5 ]na nɔ afimɛ 3Pl say PL MOD be.at here CONJ PL MOD be.at there ‘They said men should be here and women there.’ (13b) Wo be [*man wó] ne nɔ afiyi ne [*woman wó] na nɔ afimɛ (13c) Wo be [men *Ø] ne nɔ afiyi ne [women *Ø] na nɔ afimɛ The contention here is that an irregular English noun is automatically picked at the lemma level because the lemmas supporting the noun and the plural together point to it; note in the (b) examples that the singular forms of the irregular nouns are unacceptable when they are pluralized by wó. That wó is, as expected, obligatory with an irregular plural English noun is demonstrated by the unacceptability of the (c) examples from which it is omitted. 166 Legon Journal of the HUMANITIES Volume 20 (2009) Examples (14) to (18) illustrate another CS pattern that supports the explanation being offered for double plurality in the paper. In these examples the absence of wó does not lead to unacceptability: (14) Uncle be ne season dzi ɖo na cocoa la, ye-a ƒle [sandal-s a] na-m Uncle say if top reach for cocoa TP LOG-FUT buy PL the to-1sg ‘Uncle said when the season arrives for cocoa he will buy (a pair of) sandals for me.’ (Asilevi 1990: 90) (15) Me le [flower-s ya] do-ge ɖe daddy ƒe backyard garden-a me 1sg be PL this plant-INGR ALL daddy poss the in ‘I will plant this flower in daddy’s backyard garden.’ (Asilevi 1990: 23) (16) Fifia hã gbe kple ati- wó yevu wó tsɔ le tsi-mati kple [pill-s] wɔ-m Now too herb and stick-PL whitemen take be syrup and PL do-PROG mie-le zaza-m 1PL-be use—PROG ‘Now too, it is herbs and roots that the white man uses to prepare syrups and pills for us to use.’ (Nortsu-Kotoe 1999: 71) (17) Nya me se be wó be [afternoon-classe-s] gɔme dze- ge egba Evershed Amuzu 167 1sg-TP 1sg hear COMP 3PL say PL under settle-HAB today ‘As for me, I have heard that afternoon classes commence today.’ (Amuzu 1998: 73) (18) Wo forecast be [flood-s] ga nu gblẽ–ge this year 3PL COMP PL REP thing spoil-PROG ‘It has been forecast that floods will cause a lot of damage this year.’ The underlined English nouns exemplify the kind of nouns whose semantics Tiersma (1982) describes in one of his principles of “Local Markedness”. He says that When the referent of a noun naturally occurs in pairs or groups, and/or when it is generally referred to collectively, such a noun is locally unmarked in the plural. (Tiersma (1982: 835) Note that sandals (14) are entities which always occur in pairs; that flowers (15) and pills (16) are entities which are generally referred to collectively; that extra classes constitute an event which entails sub-events or individual lessons; and that floods denote plenty of water over a hitherto dry parcel of land. These nouns are “locally unmarked in the plural”, i.e. they may conveniently be regarded as singular nouns. Even though the plural –s form appears on such nouns, it does not do so because it is conceptually activated with them. And because –s is not conceptually salient as a plural at the lemma level, there are no grammatical procedures at the functional level either for the projection of the obligatory +5 slot for wó to automatically fill. Note, however, that when wó does appear with such a noun, it signals that different kinds, or different groups, of the entities in question are being referred to. For example, sandal-s a wó in (19), a version of (14), means ‘the pairs of sandals’, and flower-s ya wó in (20), a version of (15), means ‘these kinds of flower’: When the referent of a noun naturally occurs in pairs or groups, and/or when it is generally referred to collectively, such a noun is locally unmarked in the plural. (Tiersma (1982: 835) 168 Legon Journal of the HUMANITIES Volume 20 (2009) (19) Uncle be ne season dzi ɖo na cocoa la, ye-a ƒle [sandal-s a wó] na-m Uncle say if top reach for cocoa TP LOG-FUT buy PL the PL to-1sg ‘Uncle said when the season arrives for cocoa he will buy the pairs of sandals for me.’ (20) Me le [flower-s ya -wó] do-ge ɖe daddy ƒe backyard garden-a me 1sg be PL this PL plant-INGR ALL daddy poss the in ‘I will plant this flower in daddy’s backyard garden.’ In (19) and (20), we may say that the nouns are locally marked in the plural, and that wó is accordingly mandatory in the slot +5. Concluding Remarks4. This paper dealt with the phenomenon of double plurality in Ewe-English CS. It explained that while –s pluralizes English nouns in the relevant mixed NPs on semantic ground (i.e. –s is conceptually activated at the lemma level along with the English nouns), wó pluralizes those English nouns on structural ground (i.e. wó is obligatorily assigned a +5 plural slot following the application of the MOP and the SMP at the functional level). Based on this explanation, one might be tempted to suggest that there is a connection between the times of activation of the two plurals and their order of appearance in the mixed NP (i.e. that -s always comes before wó because it is selected at the lemma level while wó always follows later because it is selected at the functional level). This suggestion has no merit. I dare say that were Ewe a prefix-plural language, the two plurals would have distributed in the order plurals double in Acholi-English CS. Myers-Scotton (2006: 207) cites lu-civilian-s as a typical example of double plurality in Acholi-English CS. Here lu, the Acholi Evershed Amuzu 169 plural, is a prefix while –s as expected occurs as suffix on the English noun stem. The point is that in both Ewe-English CS and Acholi-English CS the lemma and functional level processes culminate in framing the mixed constituent surface structure. Having said so, I should revisit the 4-M model classification of plural morphemes. Following the model we have assumed matter-of-factly that the plural is an early system morpheme. There is reason to believe, however, that there is more to plurals than the 4-M model currently reveals. The model distinguishes early system morphemes from late system morphemes on the basis of two features, [± Conceptually Activated] and [± Quantification] and classifies plurals as early system morphemes. The former feature is concerned with whether a system morpheme has a semantic or pragmatic value with the content morpheme it accompanies. The latter feature, on the other hand, deals with whether a system morpheme is a critical or mandatory grammatical element in the distribution of content morphemes. Early system morphemes have a plus reading for the first feature but a minus reading for the second one. For late system morphemes, the reverse is the case (see Myers- Scotton 2002: 73). In reference to their functions in the NPs where they co-occur, it is only –s which qualifies as an early system morpheme; in those NPs wó needs to be analysed as a late system morpheme. Of course calling wó a late system morpheme is novel since plurals have hitherto not been classified as late system morphemes. But that is not the point here. The argument, rather, is that the evidence adduced in this paper points to plurals having plus settings for both features and that the phenomenon of double plurality derives directly from the fact that the two features may be split between plural forms from different languages.2 This argument has implications for our analysis of the plural category outside NPs showing double plurality. In NPs in which there is no plural doubling (i.e. in monolingual NPs and in mixed NPs in which only one plural is realized), the one plural form may be said to combine the two plural functions. Wó exemplifies such plurals. It performs the multiple functions in monolingual as well as mixed 170 Legon Journal of the HUMANITIES Volume 20 (2009) NPs—e.g. the (a) versions of (5) to (8)—where it occurs alone. In example (1), reproduced below as (21), wó occurs with student in the multiple-function capacity but with textbook in only the grammatical-function capacity because –s expresses the conceptual plural function with textbook. (21) Headmaster la inform [student 0 -aɖe +4 -wó +5 ] be wó-a- label DEF INDEF-PL COMP 3PL-SUBJ [{textbook 0 s +1 } yɛyɛ +1 -a +4 -wó +5 ]. PL new the PL ‘The Headmaster informed some of the students to label the new textbooks.’ (Asilevi 1990: 34) Wó plays the multiple functions as well as the late (i.e. grammatical) plural function in mixed NPs because it comes from the language that is active from the lemma level through to the functional level; –s, on the other hand, is restricted to playing the early (i.e. conceptual) plural function in the mixed NPs because it comes from a language that may be active only at the lemma level. Now, in order not to ignore the fact that plurals are not always purely early system morphemes, as is claimed in the 4-M model, I propose the following five- way (instead of the current four-way) classification of morphemes for further cross-linguistic verification: Content morphemes (i) (Purely) early system morphemes: e.g. the English –(ii) s in mixed NPs showing double plurality, as in (1) to (4); and prepositions in lexicalized verb phrases, e.g. up in break up Multiple-function system morphemes: e.g. most plurals in monolingual (iii) contexts; and some plurals as used in mixed NP, e.g. wó in mixed NPs not showing double plurality, as in the (a) versions of (5) to (8)3 Evershed Amuzu 171 Late bridge system morphemes(iv) (Purely) late system morphemes: e.g. (v) wó as used in mixed NPs showing double plurality, e.g. (1) to (4); and case-markers, noun number markers, etc. My last concluding remark is that there is evidence that the phenomenon of double plurality is not a universal feature of CS and that whether it obtains or not depends on the typology of the languages involved. To this end, the existence of double plurality in Ewe-English CS as opposed to its non-existence in Akan- English CS and Ga-English CS is revealing. Unlike Ewe and English which have just one plural marker each, both Akan and Ga, two languages related to Ewe which also have CS relationship with English, have four plural markers.4 It turns out, as preliminary studies show (see Quarcoo, 2009 and Agyei-Owusu 2009), that in both Akan-English CS and Ga-English CS with the exception of the profession plurals -foɔ (Akan) and -foi (Ga) only the English suffix –s may occur on an English noun in bilingual clauses. In the following examples, therefore, English nouns consistently occur with only –s (contrary to the fact that in Ewe-English CS the Ewe counterpart would automatically also appear); Quarcoo and Agyei-Owusu in their independent studies noted that the appearance of any Akan / Ga plural marker in addition to –s would make the sentences unacceptable: (22) [Saa mo farm-s no], [mo correspondent-s a ɛ- wɔ hɔ no] n-kaa ho That your PL DET your PL who 3PL- be there DET NEG-say skin asem n-kyerɛ mo? matter NEG-tell you ‘Those farms of yours, your correspondents who are there, haven’t they told you anything about them?’ (Akan-English CS; Quarcoo 2009) 172 Legon Journal of the HUMANITIES Volume 20 (2009) (23) [Problem-s no] deɛ ɛ-bɛ ba, na yɛ bɛ-solve -o PL DET EMP 3sg-FUT come, and 1PL FUT-solve ‘As for the problems, they will come and we will solve (them).’ (Akan-English CS; Quarcoo 2009) (24) Mechanic-s ɛ́ ní ò-bàá-tsɛ́ àmɛ̃-!ɛ́ PL DET that 2sg-FUT-call them-DEF ‘The mechanics that you call...’ (Ga-English CS; Agyei-Owusu 2009) (25) Mì-! nã mĩĩshɛ̀ɛ̀ yɛ̀ [proverb-s-ɛ́] ní nyɛ́-tsɔ̀ɔ́ wɔ́-!ɔ́ 1sg-get happiness LOC PL-DET that 2sg- teach us-DEF ‘I am happy that you have taught us these proverbs.’ (Ga-English CS; Agyei-Owusu 2009) It seems to me that after –s is selected along with the English nouns during lemma level processes no fresh processes are initiated at the functional level to project an obligatory slot for an Akan / Ga plural, because there are competing candidates. In fact, as noted, with the exception of the group / profession plurals -foɔ (Akan) and -foi (Ga), no Akan/Ga plural marker may even occur alone with an English noun. This is illustrated with the following experiment with the English nouns brother and teacher (note that the profession markers go with teacher because teachers do constitute a professional group): Akan: *a-brother, *n-brother, *brother-nom, *brother-foɔ ‘brother’ *a-teacher, *n-teacher, *teacher-nom, teacher-foɔ ‘teachers’ Evershed Amuzu 173 Ga: *brother-ji, *brother-i, *brother-bii, *brother-foi ‘brothers’ *teacher-ji, *teacher-i, *teacher-bii, teacher-foi ‘teachers’ This situation, where foreign-origin nouns are accorded a different morphological treatment from what is accorded their indigenous counterparts, is not extraordinary in language contact. It is well-known, for example, that foreign nouns borrowed or switched into noun-class languages are (usually) not subjected to the noun classification system in the borrowing languages; they are generally assigned to one or two noun classes irrespective of the classes to which their counterparts in the borrowing languages belong. This has happened in noun borrowing into Swahili and Logba (see Myers-Scotton 1993, 2002 and Dorvlo forthcoming respectively), and we may conjecture that the phenomenon is a simplification strategy by which the borrowing speakers avoid having to apply the intricate language-internal noun-class system to borrowed nouns. I believe that a similar strategy is at play in Akan-English and Ga-English CS with respect to the pluralization of switched English nouns: the English nouns are exempted from inflecting for any of the four Akan/Ga plurals, an exemption which spares the Akan/Ga codeswitcher from worrying about plural form which suits an English noun he/she wants to use in CS. Consequently, English nouns that require plural marking in CS contexts occur with only the English –s. Further cross-linguistic CS research may confirm the following CS typology with regard to plural marking: When the language of functional level processes (i.e. the language • of CS grammar) has more than one plural marker (as is the case with Akan and Ga), plural doubling would/may be inhibited as a result of a simplification strategy to avoid applying a language-internal plural selection system to switched/borrowed nouns. 174 Legon Journal of the HUMANITIES Volume 20 (2009) When, on the other hand, the language of CS grammar has just one • plural marker (as is the case with Ewe), plural doubling would be optional and the distribution of the two plurals would follow the rules that have been spelt out in this paper. Evershed Amuzu 175 NOTES For a clarification of how slots are numbered in mixed NPs, see Appendix 2 below.1. We have repeate2. dly implied that –s performs the conceptual (early system) plural function and that wó performs the grammatical (late system) plural function in the mixed NPs in which they co-occur. And, we have also repeatedly implied that the splitting of the two functions results from the fact they do not become salient at the same stage during language production. Some deter3. miners, demonstratives, and intensifiers may also be found to be multiple- function system morphemes. In Akan, there are 4. at least four plurals: the homorganic nasal plural prefix n-, another prefix a-, and two suffixes: -foɔ, which is a group or profession plural marker, and nom, which is a kinship plural marker. Ga also has four plural suffixes: -ji, -i, -bii, and the group or profession plural suffix foi (Agyei-Owusu, 2009). 176 Legon Journal of the HUMANITIES Volume 20 (2009) APPENDIX 1 In the following codeswitching examples of Ewe-English CS, notice that the following grammatical categories have been expressed two times, first in English and then in Ewe: the determiner (1) and (2), the proximal demonstrative (3), and the intensifier (4): (1) Me dzi be na tsɔ [{the usual type} -a ko] 1sg want that 2sg.SUBJ take the only a-gbã xɔ-a FUT-roof house-the ‘I would like you to use only the usual type to roof the house.’ (Amuzu 1998: 74) (2) E-kpɔ be [Azumah ƒe {style thoughout the fight} –a] 2sg-see that Azumah poss the me nɔ hectic abe previous fights- wó o NEG beNPRES like PL NEG ‘You’ll observe that Azumah’s style throughout the fight wasn’t as hectic as previous fights. (Asilevi 1990: 59) (3) Be ma set example a, [{this -4 Saturday 0 } ya +4 ko +7 ] nye cousin… COMP 3sg-POT TP, this only 3sg ‘To set an example, it is only this Saturday that my cousin….’ (4) Xɔ-a le [ {very beautiful ŋutɔ} ] House-the be very ‘This house is very beautiful.’ Evershed Amuzu 177 APPENDIX 2 For the purpose of this analysis, and following Duthie (1996), I refer to the slot in which a noun functions as head of an NP as 0 slot (“zero slot”), the “nucleus” of the NP. The 0 slot is the “centre of gravity” in relation to which we have pre- and postposed peripheries – i.e. pre-modifiers and post-modifiers respectively. Each modifier slot is identified by a plus / minus digit that reflects how many places it is to the right or to the left of the 0 slot. Thus, the English plural is marked +1 to show that its slot is directly to the right of the head. As will become clear, the order of Ewe morphemes in relation to the English noun head is what we find in monolingual Ewe NPs. A comprehensive outline of the Ewe NP structure is therefore provided in the table below. Notice, for example, that the Ewe plural wó is in +5 slot. 178 Legon Journal of the HUMANITIES Volume 20 (2009) -2 -1 0 +1 +2 +3 +4 +5 +6 +7 +8 po r-N P (IN T1 ) N PR O (A D J) (C A RD ) (O R D ) (D ET ) (R EL ) (P L) (Q T) (IN T2 ) (C LS ) K of i ƒ e K of i’s éƒ e ‘h is’ ny e ‘m y’ ne ne m , a le sig be su ch N yu i ‘g oo d’ lo lo ‘b ig ’ bu bu ‘o th er ’ ɖe ka ‘o ne ’ ev e ‘tw o’ Ev el ia ‘s ec on d’ bú bu ‘o th er ’ m am lɛ ‘la st, re m ai n- in g’ -(l )a ‘t he ’ aɖ e ‘s om e’ m a ‘th at ’ si w hi ch w ó ka ta  ‘a ll’ pɛ / k o ‘o nl y’ hã ‘t oo ’ bo ŋ ‘ra th er ’ ko ŋ ‘in pa rti cu la r’ T he E w e N ou n P hr as e S tr uc tu re ( A m uz u 20 05 a) Evershed Amuzu 179 REFERENCES Agyei-Owusu, Akua. (2009). The MLF Model and Ga-English Codeswitching. An M.Phil Sociolinguistics Term Paper, Department of Linguistics, University of Ghana. Amuzu, Evershed K. 1998. Aspects of Grammatical Structure in Ewe-English Codeswitching. M. Phil. Thesis. Oslo: University of Oslo. Amuzu, Evershed K. 2005a. Ewe-English Codeswitching: A Composite rather than Classic Codeswitching. PhD. Dissertation, Australian National University. Amuzu, Evershed K. 2005b. Revisiting the Classic Codeswitching – Composite Codeswitching Distinction: A Case Study of Nonverbal Predication in Ewe-English Codeswitching. In M. Florey and P. McConvell (eds.), Australian Journal of Linguistics 25.1, 2005, pp. 127-151. Amuzu, Evershed K. Forthcoming. Producing Composite Codeswitching: The role of the modularity of language production. International Journal of Bilingualism. Asilevi, Francis. 1990. English-Ewe Code-mixing in Conversational Discourse: A case of English as a second language in Ghana. MA Thesis. Accra: University of Ghana. Dorvlo, Kofi. Forthcoming. Ewe Borrowings into Logba. International Journal of Bilingualism. Duthie, A. 1996. Introducing Ewe Linguistic Patterns. Accra: Ghana Universities Press. Grosjean, Francois. 2001. The Bilingual’s Language Modes. In Janet Nicol (ed.), One Mind, Two Languages: Bilingual Language Processing. Oxford: Blackwell.Myers-Scotton, Carol. 1993. Duelling Languages: Grammatical Structure in Codeswitching.Oxford: Clarendon Press. Myers-Scotton, Carol. 2002. Contact Linguistics: Bilingual Encounters and Grammatical Outcomes. Oxford University Press. 180 Legon Journal of the HUMANITIES Volume 20 (2009) Myers-Scotton, Carol. 2006. Natural codeswitching knocks on the laboratory door. Bilingualism: Language and Cognition 9 (2), 203-212. Myers-Scotton, Carol and Jake, Janice. 2001. Explaining Aspects of Codeswitching and Their Implications. In: Janet Nicol (ed.), One Mind, Two Languages: Bilingual Language Processing, 84-116. Oxford: Blackwell. Nortsu-Kotoe, Peter, K. 1999. A Linguistic Analysis of Ewe Adverts. BA Long Essay. Accra: University of Ghana. Quarcoo, Millicent. 2009. Grammatical Constriants on Noun and Verb Phrases in Twi-English Codeswitching. MPhil Thesis. Legon: Department of English, University of Ghana. Tiersma, Peter Miejes. 1982. Local and General Markedness. Language, Vol. 58-4. 181 Comparing Demonstratives in Kwa Nana Aba Appiah Amfo and Stella Boateng Norgbe1 Abstract This paper is a comparative study of demonstrative forms in three Kwa languages, Akan, Ga and Ewe. It examines the syntax, morphology, semantics and pragmatics of the demonstrative systems in these languages. The existence of various categories such as demonstrative determiner, demonstrative pronoun, demonstrative adverb, demonstrative identifier and the under-described ‘particular’ demonstratives in these three languages is examined. Semantically, we observe a basic two-way deictic contrast for most of the demonstrative systems, even though there are a few variations. The most significant qualitative feature distinction observed is one of humanness. The dearth of the morphology of these forms is noted. Finally, the role of the demonstratives in the organisation of speech flow is considered. The description of the demonstrative systems of these languages could lead to an eventual re-construction of the demonstrative system in the proto-language. Introduction1. 1 Demonstratives have received various characterisations from different authors, however what is common to all these characterisations is the deictic feature associated with demonstratives. Lyons (1977) describes demonstratives as referring expressions which provide information about the location of an object in relation to a deictic centre. Dixon (2003: 61-62) suggests that a demonstrative is “any item, other than 1st and 2nd person pronouns, which can have pointing or (deictic) reference”. Diessel (1999), on the other hand, provides a more extensive characterisation of a demonstrative based on syntactic, semantic and pragmatic features. We will look at these features, as suggested by Diessel, in turn. Syntactically, demonstratives are deictic expressions that serve specific syntactic functions. These include pronouns, adnominals and locational adverbs. Semantically, demonstratives may be identified with some specific semantic features. The most prominent of such semantic features is the distance distinction that characterises many demonstrative systems, even though some 1 Dr. Nana Aba Appiah Amfo is a Senior Lecturer in the Department of Linguistics, University of Ghana, Legon; and Mrs. Stella Boateng Norgbe is a Tutor in the Language Centre also of the University of Ghana, Legon. 182 Legon Journal of the HUMANITIES Volume 20 (2009) languages distinguish demonstratives on the basis of other qualitative features such as animacy, humanness, and number. A few languages, such as French (ce/cette/ ces), have distance-neutral demonstratives. Typically however, most languages indicate a two-way distinction; they have proximal demonstratives which denote closeness to the deictic centre and distal demonstratives which denote some relative distance from the deictic centre, such as English this and that, here and there. Some languages tend to be more elaborate in their distance distinction. Spanish, Portuguese and Tamil are among languages which make a three-way distinction. For instance, Spanish distinguishes between the locational adverbs aqui ‘here’, ahi ‘just there’ and alli ‘over there’. Northern Sami, Samal and Hausa are characterised with four-way distinctions. Jagar (2001) reports that Hausa has four Locational adverbs: nân ‘here, near the speaker’, nan ‘there, near the addressee’, cân ‘there, away from speaker and addressee’ and can ‘there, further away from speaker and addressee’. Even more elaborate systems can be found in Malagasy (Austronesian), six-way distinction; Daga (Trans-New Guinea), fourteen-way distinction; Yup’ik (Eskimo-Aleut) thirty-way distinction (Anderson and Keenan 1985). However, more elaborate distinctions tend to include other features such as (non)visibility, uphill/downhill, lower/higher elevation, upriver/downriver. Pragmatically, demonstratives are used to organise information flow in an ongoing discourse by keeping track of prior discourse participants and activating shared information. Mostly, the referents of demonstratives are ‘activated’ or at least ‘familiar’ (cf. Gundel et al 1993). Their main purpose is to focus the addressee’s attention on objects or locations in the speech situation (Diessel 1999). Following Amfo (2007), this paper aims at making a contribution to our understanding of demonstratives in three Kwa languages spoken in Ghana, Akan, Ewe and Ga. Even though these three languages are among the most prominent local languages in Ghana, little is known about their demonstrative systems2. The paper seeks to describe the types and forms of demonstratives in these languages in a comparative perspective. Attention will be paid to the morphological make- up of these demonstratives. The syntactic positions of the various categories of demonstratives will be analysed. Finally, the semantics and pragmatics of the demonstratives will be examined. We will focus on the deictic and qualitative features embedded in the demonstrative forms. Of particular interest will be what we suggest to be a grammaticalization process involving the Ewe proximal demonstrative 183 determiner sia. This paper is a step in the direction of providing a typological description of the demonstrative systems in Kwa, beginning with the select Ghanaian languages. The rest of this paper is organised as follows. The next brief section 2 provides some language information about the languages in question. Section 3 focuses on the types of demonstratives present in the languages. Diessel’s categories of demonstratives are introduced and these form the basis of the specific language analyses of Akan, Ewe and Ga in subsections 3.1, 3.2 and 3.3 respectively. Attention is paid to the various syntactic positions of the different demonstrative forms. Section 4 is devoted to the rather generally invariable morphology of the demonstratives, while section 5 discusses the deictic and qualitative semantic features embedded in these demonstratives. Section 6 is devoted to some pragmatic features of the demonstratives, and it includes an examination of the grammaticalisation process involving the Ewe proximal demonstrative determiner sia. Section 7 is the conclusion. Language Information2. Akan, Ewe, Ga are all Kwa languages belonging to the Niger-Congo language family. Ga belongs to the Ga-Dangme sub-group, a branch of the Nyo group, whereas Ewe belongs to the Gbe sub-group, which descends from the Left Bank group. Akan is a direct descendant of the Central Tano sub-goup (cf. Stewart 1989). Akan is the most prominent Ghanaian language,3 with about 40% of the total population considered as native speakers, and about two-thirds of the population use it as a medium of communication for social, economic and religious purposes (Obeng 2005). The analyses made in this paper are based on the three literary dialects: Akuapem, Asante and Fante. As a result, we draw examples from these three dialects, and they are indicated accordingly AK, AS and FA for Akuapem, Asante and Fante respectively. Ewe is the second largest indigenous language in Ghana.4 It is used as a cover term for a number of dialects including Anlo, Peki, Kpando, and Gbi (Duthie 1988)5. Native speakers of Ewe occupy most parts of the Volta region. Ewe serves as a lingua franca for native speakers of some Potou Tano languages such as Lelemi, Avatime, Logba, as well as Guan languages like Nkonya. Unless otherwise indicated, our examples are drawn from the standard variety, which incorporates features of a number of dialects including Peki, Amedzofe and Keta. 184 Legon Journal of the HUMANITIES Volume 20 (2009) Ga is the native language of people from in and around the capital of Ghana, Accra. It is spoken by a number of non-Gas who have lived and possibly grown up in Accra.6 Dakubu (1988) reports that there is little regional variation in the language, and there are no clearly distinguishable dialects. Types of Demonstratives3. The categorisation of demonstratives in Akan, Ewe and Ga done in this paper, is based, to a large extent, on Diessel’s (1999) classification of demonstratives. Diessel identifies four categories of demonstratives: demonstrative pronoun, demonstrative determiner, demonstrative adverb and demonstrative identifier. A demonstrative pronoun is a demonstrative which occurs independently in argument positions of verbs and it represents a noun, as illustrated in (1). On the other hand, a demonstrative determiner co-occurs with a noun in a noun phrase, as shown in (2). Janet prefers 1) this. 2) That boy is my son. A demonstrative adverb functions as a verb modifier. As expected, they are found in argument positions of verbs, as illustrated by English here in (3). Freda died 3) here. Diessel identifies two kinds of demonstrative adverbs: Locational deictics and manner demonstratives. Following Amfo (2007), we will label them Locational demonstrative adverb (LDA) and Manner demonstrative adverb (MDA). LDAs are “primarily used to indicate the location of the event or situation that is expressed by a co-occurring verb”. MDAs “involve some comparison and they are often used as discourse deictics” (Diessel 1999: 74). They are often glossed as ‘in this/ that way’ or ‘like this/that’. A demonstrative identifier is a demonstrative which occurs in a specific grammatical construction, a copula or non-verbal clause. Their communicative function is to focus the addressee’s attention on entities in the extra-linguistic context. Example (4) involves a demonstrative identifier nie from Akan. Mpaboa no 4) nie.7 Nana Aba Appiah Amfo and Stella Boateng Norgbe 185 Shoes DEF DI ‘Here are the shoes.’ There is a fifth category of demonstratives identified in the Kwa languages examined here, which is not captured under Diessel’s categorisation. In Akan and Ga, the demonstratives in this category are identical with the MDAs in the respective languages. In Ewe, there is a phonological similarity between the two forms. Following Schuh (1977), we call these ‘particular’ demonstratives. These demonstratives are adnominal, but unlike other nominal modifiers in the languages, they occur pre-nominally, and they obligatorily require the presence of a post- nominal determiner as well. The utterance in (5) involving the Akan ‘particular’ demonstrative saa is ungrammatical without the post-nominal determiner yi. 5) Saa abofra yi nim adeε paa. PD child PDD know thing INT ‘This child is very intelligent.’ The following sub-sections discuss the categories of demonstratives in the various languages in more detail. Akan3.1 All four categories of demonstratives cited in Diessel have been identified in Akan. In addition, as indicated in the previous section, there is a fifth category. What follows is a summary of the types of Akan demonstratives (cf. Amfo 2007) and their various syntactic positions. There are two identifiable demonstrative determiners in Akan: yi and no. Yi is the proximal demonstrative, denoting closeness to the speaker, and no can be considered as its distal counterpart denoting relative distance away from the speaker. In Akan not only do demonstrative determiners occur as nominal modifiers, they are used as clausal modifiers as well. They may occur in post- nominal position when they modify a noun, otherwise they occur clause finally, taking scope over the whole clause, as illustrated in (6) and (7) respectively.8 Asem 6) yi yε-ε ɔbarima no ne ne yere no Matter PDD do-COMPL man DEF CONJ POSS wife DEF dε mmoroso. 186 Legon Journal of the HUMANITIES Volume 20 (2009) sweet overflow ‘The man and his wife were exhilarated by this message.’ AK Me-te-ase 7) yi, me-m-pene m-ma Agyeman n-ware I-stay-under PDD, I-NEG-agree NEG-allow Agyeman NEG-marry Kyeiwaa. Kyeiwaa ‘As long as I am alive, I wouldn’t agree to Agyeman and Kyeiwaa’s marriage.’ AS The demonstrative pronouns in Akan are the proximals eyi/oyi/iyi/wei and the distals ɛno/ɔno. Akuapem uses eyi for non-human references and oyi for human ones. Asante uses wei for both human and non-human references, and Fante iyi is also underspecified in terms of humanness. In Asante and Akuapem, ɛno refers to non-human entities, whereas ɔno is used in reference to human entities. Like its proximal counterpart, Fante ɔno is used in reference to both human and non- human entities.9 Demonstrative pronouns in Akan occur in argument positions of verbs. Specifically, they occur in subject, object and focus positions, as exemplified in (8), (9) and (10) respectively. 8) Iyi pε mpuwa, na iyi pε ekutu. PDP like bananas, CONJ PDP like oranges ‘This one like bananas and this one like oranges.’ FA Me-t9) ɔ-ɔ iyi wɔ Kumase. I-buy-COMPL PDP be.at Kumase ’I bought this one at Kumase.’ FA Nana Aba Appiah Amfo and Stella Boateng Norgbe 187 10) Iyi na me-hu-u no wɔ fie hɔ no. PDP FM I-see-COMPL him be.at house t h e r e DEF ‘This person is the one I saw there at the house.’ FA Notice that when the pronoun occurs in focus position, the main clause contains a resumptive pronoun which is co-referent with the demonstrative pronoun, the exception being when the referent of the demonstrative is an inanimate object. There are two kinds of demonstrative adverbs in Akan, as identified in Amfo (2007). They are Locational demonstrative adverbs (LDA) and Manner demonstrative adverb (MDA). The LDAs are (ε)ha ‘here’ and (ε)hɔ ‘there’. There are four syntactic positions associated with the Akan LDAs. First, they may occur as adverbial complements in object position. In this position, they can occur independently, or they may co-occur with other adverbials, as seen in (11). Ye-sie-e no 11) ha (nnεra). We-bury-COMPL him/her here (yesterday) ‘We buried him/her here (yesterday).’ AS Second, they may occur as head nominals in noun phrases contained in constructions which make reference to parts of the body, as illustrated in (12). Such utterances have to be accompanied with pointing gestures. Me 12) ha/hɔ a-hono. My here/there PERF-swell ‘Lit.: My here/there is swollen’ (This/that part (of my body) is swollen.) Third, they may function as the modifier of a head noun. In (13) εha ‘here’ modifies sukuufoɔ‘pupils’, and it indicates that the pupils of an area close to the deictic centre do not like learning. Of course, whether the demonstrative εha 188 Legon Journal of the HUMANITIES Volume 20 (2009) ‘here’ is used in reference to an institution, a town or even country is subject to pragmatic inference. Ɛh13) a sukuufoɔ m-pε adesua. Here school.people NEG-like learning ‘Pupils here do not like learning.’ (e.g. Pupils of this institution/town do not like learning.) Finally, the LDAs may be found in complex noun phrases which involve a head noun, a demonstrative determiner and a postposition. In such constructions, the proximal demonstratives (i.e. determiner and LDA) co-occur, as shown in (14), while the distals no and ha collocate. Kuro yi mu 14) ha deε, wo-n-nya yere. Town PDD inside here TM, you-NEG-get wife ‘As far as this town is concerned, you will not get a wife (from here).’ AS Akan has deictically contrastive MDAs. They are dεm and dei for Fante and saa and sei for Akuapem and Asante.10 Sei/dei is the appropriate MDA when the speaker coincides with the demonstrator of the action being performed, and the action and the utterance are performed simultaneously, or the action immediately follows the utterance. Saa/dεm is the felicitous MDA when the action is performed before the saa-sentence is uttered, or when it is someone else other than the speaker who is performing the action being referred to, or else it is used discourse anaphorically in reference to some already mentioned state of affairs. We can thus consider sei/dei as the proximal form and saa/dεm its distal counterpart. The MDA has two syntactic positions; it either occurs in the object position, or else it is found in the focus position, but never in subject position. In (15), sei is apt when the utterance is accompanied by the appropriate gestures performed by the speaker. Saa may be used either discourse anaphorically or when someone else than the speaker is doing the gesturing. In (16), Fante dεm may be used discourse Nana Aba Appiah Amfo and Stella Boateng Norgbe 189 anaphorically, or else it may be used together with the suitable gestures by some other interactant who is not the speaker. 15) Saa/Sei na o-didi. MDA FM s/he-eat.RED ‘That is the way s/he eats.’ AS Me-m-p16) ε dεm. I-NEG-like MDA ‘I don’t like that’. AS Akan has another demonstrative which does not seem to fit any of the categories suggested by Diessel, and which has not received an adequate description so far. This may be due to its phonologically identity with the distal MDA, and its lack of independent occurrence, in that it always occurs with a demonstrative determiner. Indeed, it is ungrammatical to have the pre-head modifier saa/dεm without a post-head demonstrative determiner. Its syntactic position is also distinct; unlike most noun modifiers in the language, it is a pre-head modifier. Example (17) illustrates the use of this demonstrative. 17) Saa papa no/yi ba-a ha. PD man DDD/PDD come-COMPL here. ‘That man came here.’ This category of demonstratives has not received wide attention cross-linguistically. However, Schuh (1977) reports of such a category of demonstratives in the Ngizim/ Bade (Afroasiatic, Chadic) languages that he studies. He labels it as a ‘particular’ category, and admits the relative difficulty in precisely categorising this group. Following Schuh, we call the pre-nominal saa/dεm a ‘particular’ demonstrative. It could be translated as ‘that same …’ or ‘that particular …’. It refers to a specific entity which is ‘familiar’ to both speaker and addressee (in which case it combines with no), or ‘activated’ (in which case it combines with yi). The use of this demonstrative signals the specificity of a familiar or an activated referent. 190 Legon Journal of the HUMANITIES Volume 20 (2009) Demonstrative identifiers, as have already been indicated, occur in specialized constructions. In Akan, demonstrative identifiers occur in non- verbal clauses. They occur together with noun phrases. The noun phrase may be a very simple one, which contains only a proper noun, or a head noun and a determiner, to more complex ones, such as noun phrases containing modifiers like adjectives or numerals, to even more complex noun phrases such as those with clause size modifiers like relative clauses. The Akan demonstrative identifier is ni(e) (Akuapem/Asante) and nyi (Fante), and the different syntactic collocations outlined above are illustrated in (18) to (21) below. 18) Agyeiwaa nie. Agyeiwaa DI ‘Here is Agyeiwaa’ AS N-ky19) εnsee no nie. PL-pan DEF DI ‘Here are the pans.’ AS M-maa mmienu no 20) nie. PL-woman two DDD DI ‘Here are those two women.’ AS Abofra no a o-di-i aduane no 21) nie. Child DEF REL s/he eat-COMPL food DCM DI ‘Here is the child who ate the food.’ Ewe3.2 Demonstrative determiners, demonstrative pronouns, demonstrative adverbs, demonstrative identifier and ‘particular’ demonstrative are the categories present in Ewe. The demonstrative determiners are the distal ma and the proximal sia. Nana Aba Appiah Amfo and Stella Boateng Norgbe 191 These demonstratives are adnominal, they may co-occur with a head noun with or without modifiers, as illustrated in (22) and (23) below, or they may occur in more complex noun phrases where there are other modifiers. The utterance in (24) contains an adjective in addition to the object head noun. Agbale 22) sia nyo. Book PDD be.good ‘This book is good.’ Me-xle agbale 23) ma kpɔ I-read book DET PERF ‘I’ve read that book.’ 24) Me-xle agbale xɔåkɔ ma kpɔ. 1-read book collect-name DET PERF ‘I’ve read that famous book.’ Demonstrative pronouns in Ewe are formed by prefixing the determiners with the third person singular pronoun e. The result is esia ‘this one’ and ema ‘that one’. The pronouns occur independently in argument positions of verbs as in example (25) and (26). Their referents may be concrete entities which are visible in the extra linguistic context, as may well be the case in (25), where the speaker points to two objects to show the distinction between esia (this one) and ema (that one), or else the pronouns are used anaphorically in reference to an earlier discussed situation as illustrated by example (26). In this particular example, esia is used to refer to a situation where citizens of a certain village were always falling sick due to the fact that they had refused to keep their environment clean, a situation which had already been mentioned in an earlier utterance in the course of the ensuing discourse. Esia nyo wu 25) ema. PDP good surpass DDP ‘This one is better than that one.’ 192 Legon Journal of the HUMANITIES Volume 20 (2009) 26) Ameka ∫e vodada e-nye esia. Who POSS fault PRO-be PDD ‘Whose fault is this?’ Unlike Akan, Ewe demonstrative pronouns are hardly used to refer to human beings. They are used in reference to inanimate and animate non-human entities. When used in reference to a human entity, it has derogatory pragmatic implications and it is clearly uncomplimentary. We have already mentioned that the demonstrative pronouns are formed by adding the 3rd person pronoun e to the demonstrative determiners. This pronoun is, strictly speaking, underspecified in terms of animacy and humanness. However, as we have indicated, the demonstrative pronouns themselves would normally not have a human referent. To express a thought such as the following: ‘I’m going home with this (person)’, one has to use the human generic term ame, which will be the head noun in that noun phrase, and it will have to be modified by a determiner sia, and not esia which is a pronoun and does not occur with the support of a noun, as shown in (27). Mayi kple ame 27) sia/*esia I.FUT.go with person PDD ‘I will go with this person.’ Ewe has three LDAs, afisia ‘here’, afima ‘there’, and afimâ ‘way over there’. Afisia ‘here’ is used to refer to an area close to the speaker, and afima ‘there’ refers to a location relatively distant from the speaker or even to an area which cannot be located in the immediate environment. The third LDA afimâ ‘way over there’ indicates a location further away from the speaker and the addressee than afima ‘there’. These LDAs are made up of two morphemes; a generic noun afi ‘place’ and the proximal and distal demonstratives determiners sia and ma for the proximal and the distal LDAs respectively.11 These demonstratives may occur as adverbial complements to a verb. In that position, it may occur on its own, as in (28), or it may collocate with other object complements. In (29), afima ‘there’ co-occurs with the time adverbial etsɔ ‘tomorrow’ as complements to the verb yi ‘to go’. As example (30) illustrates, the LDAs may also function as the object complement of the locative copula le ‘to be at’. Mi-va t28) ɔ ɖe afi-sia. Nana Aba Appiah Amfo and Stella Boateng Norgbe 193 You-come stop at place-this ‘(You) come and stand here.’ Mi-a-yi 29) afi-ma etsɔ. We-FUT-go place-that tomorrow ‘We will go there tomorrow.’ Nye-e de sukuu le 30) afi-ma. I-FM go.PST school be place-that ‘It is I who went to school there.’ Like Akan, the LDAs may occur in a possessive nominal phrase as the head noun, and this normally refers to a part of the body. These types of sentences occur only in colloquial spoken language where the various dialectal variants of the LDAs are used. This must be accompanied by the speaker or the addressee touching that part of the body. The choice between the proximal or the distal LDA depends on who is doing the touching. When the speaker is doing the touching whether on her own body or the addressee’s body, she will use the proximal LDA, on the other hand, the addressee in referring to that part of the body which is being touched will use the distal counterpart. The utterance in (31) is an example from the Peki dialect. 31) Amu-fii / fimi le venye. My-POSS here / there be pain-me ‘My here / there is hurting.’ One other characteristic of spoken colloquial Ewe is that it is possible to find the LDAs functioning as noun modifiers, in which case they syntactically precede the head noun. In (32), the speaker by modifying nyɔnuwo ‘women’ with the distal LDA afima indicates that women of some contextually determined place away from the deictic centre do not make good wives. 32) Afi-ma nyɔnu-wo me-ɖe-a srɔ o. Place-that woman-PL NEG-marry-HAB spouseNEG ‘Women from that place don’t make good wives.’ 194 Legon Journal of the HUMANITIES Volume 20 (2009) (Anlo) Ewe, has the manner demonstrative adverbs nenema/sigbe12 and ale. Nenema/sigbe is the proximal form, functioning similarly to Akan sei, and ale is the distal form comparable to Akan saa. In (33) nenema or its dialectal variant sigbe is possible if the speaker is pointing to a demonstration by a third person, who is neither the speaker nor the addressee. On the other hand, if the speaker is the same person doing the demonstration at the same time as the sentence is uttered, then ale is the felicitous demonstrative to use. Tr33) ɔ ale/sigbe/nenema. Turn like.this/that ‘Turn like this/that.’ Nenema/sigbe may be used discourse anaphorically to refer to a situation that has been described earlier in a discourse. Example (34) will be appropriate in a context where the addressee might have narrated to the speaker some experience she has had with the referent of the third person singular subject pronoun wo-. 34) Nenema/sigbe wole. Like.that s/he.be ‘That is how s/he is.’ Ewe, like Akan, has a single ‘particular’ demonstrative nenem. It is almost identical to the MDA, except that it has no final vowel ‘a’, which characterises the MDA. It occurs in pre-nominal position, co-occurring with the post-nominal demonstrative determiner ma. Unlike in Akan where the ‘particular’ demonstrative co-occurs with both demonstrative determiners, the Ewe MDA can only collocate with the distal demonstrative determiner. The referent of the nenem-NP is always activated (cf. Gundel et al), that is, it is in the current short-term memory of the addressee, possibly because it has been mentioned in the immediately preceding discourse, or it is present in the immediate extra-linguistic context. In (35), by using the pre-nominal determiner nenem and the distal demonstrative ma in the noun phrase, nenem nyɔnu … ma ‘that same … woman’, the story teller indicates to the reader that the woman he is referring to is the same woman which has already been introduced, and indeed she has been mentioned consistently throughout the story. This helps to narrow down the search area for the referent of the NP. Nana Aba Appiah Amfo and Stella Boateng Norgbe 195 35) Nenem nyɔnu kɔkɔ ma si kplɔ ɖevi-a tso PD woman tall DDD REL lead child-DEF from ave-a me la bu zideka. forest-DEF inside DCM disappear suddenly ‘That tall woman who lead the child from the forest disappeared suddenly.’ Dropping nenem out of the subject NP in (35) will suggest that the referent of that NP is ‘familiar’, rather than ‘activated’. The addressee already has a representation of the woman being talked about in his long- or short-term memory; depending on how long ago she has been mentioned or perceived. Ewe has one demonstrative identifier, esi. It occurs in copula clauses, and it is used to focus the addressee’s attention on some physically present entity. The noun phrase in such constructions could be as simple as containing a noun head and a determiner, as in (36), or it could be a more complex noun phrase with modifiers such as a relative clause as in (37). Devi-a e-nye 36) esi. Child-DEF FM-be DI ‘Here is the child.’ Agbale si dim mie-le la e-nye 37) esi. Book REL search.PROG we-PROG D C M FM-be DI ‘Here is the book that we are looking for.’ It is worth mentioning that some dialectal variants of the demonstrative identifier can be used in non-verbal clauses, as is the case in Akan. Such constructions are restricted to colloquial settings. However, what distinguishes these Ewe constructions from the Akan ones is that in Ewe the NP bears a focus marker. The following examples (38) and (39) from the Peki and Anlo dialects respectively illustrates this point. 196 Legon Journal of the HUMANITIES Volume 20 (2009) 38) Kofi-e ke. (Peki) Kofi-FM DI ‘Here is Kofi.’ 39) Srô-nye-e yia (Anlo) Spouse-POSS-FM DI ‘Here is my wife/ husband.’ Ga3.3 The categories of demonstratives found in Ewe and Akan are the same ones found in Ga. Ga has a proximal demonstrative determiner nεε, the definite determiner lεfunctions as a distal demonstrative determiner especially when it co-occurs with the pre-nominal ‘particular’ demonstrative nakai. These are illustrated in (40) and (41) below. K40) ε fɔ okplɔ nεε nɔ. Take put table PDD on ‘Put it on this table.’ K41) ε fɔ nakai okplɔ lε nɔ. Take put PD table DET on ‘Put it on that table.’ The demonstrative pronouns are the proximals enε and mεnε and the distals no and lε. The pronouns enε and no are used for non-human entities, while mεnε and lε are reserved for human entities, as we see in (42) and (43). Mi-sum42) ɔ mεnε. I-like this.person ‘I like this person.’ Nana Aba Appiah Amfo and Stella Boateng Norgbe 197 Mi-sum43) ɔ no. I-like that.one ‘I like that one.’ Ga has two LDAs and two MDAs. The LDAs are biε ‘here’ and jεmε ‘there’, and they may occur as adverbial complements in object position, sometimes in conjunction with other adverbs, as seen in (44) and (45). Mi-ba 44) biε nyε I-come.PST here yesterday ‘I came here yesterday.’ Maya 45) jεmε wɔ. I.FUT there tomorrow ‘I will go there tomorrow.’ Like their Akan and Ewe counterparts, they are sometimes used as head nominals in reference to parts of the body, as illustrated in (46). M-46) biε e-fuu My-here PERF-swell ‘Lit: My here is swollen.’ Ga, like Ewe and the Asante and Akuapem dialects of Akan, has two MDAs: proximal nεkε and distal nakai. When used with accompanying gestures, nεkε is the one used when the speaker is the one demonstrating how an action is (to be) performed, while nakai is used when the speaker is not the one demonstrating, but she is pointing at another person who is performing the action. Thus, in (47), the choice between nεkε and nakai depends on whether the demonstrator coincides with the speaker or not. Fee-m47) ɔ lε nεkε/nakai. Do-NOM PRO like.this/like.that ‘Do it like this/that.’ 198 Legon Journal of the HUMANITIES Volume 20 (2009) Not surprisingly, nakai is the MDA that can be used anaphorically in reference to some previously described action or behaviour. ‘Particular’ demonstratives, identical with the MDAs, are present in Ga. Like their Akan and Ewe counterparts they collocate with post-nominal determiners. The proximal ‘particular’ demonstrative nεkε combines with the proximal demonstrative determiner nεε as in (48), and nakai, the distal ‘particular’ demonstrative is used together with lε, as example (49) shows. 48) Nεkε gbekε nεε e tue wa. PD child PDD he ear hard ‘This child is stubborn.’ 49) Nakai gbekε lε, e gbo. PD child DEF, he die.PST ‘That child is dead.’ What is significant with the Ga ‘particular’ demonstrative constructions, which is absent in Akan and Ewe, is the presence of resumptive pronouns. Both examples (48) and (49) contain the resumptive pronoun e, which is co-referent with the noun phrase meaning ‘this/that child’. The demonstrative identifier in Ga is nε. Like in Akan and in some dialects of Ewe, the demonstrative identifier occurs in non-verbal clauses, as example (50) shows.13 Mi-hefatal50) ɔ nε. My-spouse DI ‘This is my spouse.’/ ‘Here is my spouse.’ Again, comparable to the other languages we have talked about, the noun phrases in such non-verbal clauses have no limitations on their complexities. Example (51) includes a noun phrase which contains a noun modifier, a numeral and a determiner in addition to the head noun, and (52) has a head noun which is modified by a relative clause. Nana Aba Appiah Amfo and Stella Boateng Norgbe 199 Gbek51) ε-bii hii etε lε nε. Child-PL boys three DEF DI ‘Here are the three boys.’ Nuu l52) ε ni fa shika lε yε mi-dεŋ lε nε. Man DEF REL borrow money DEF at my-hand DCM DI ‘Here is the man who borrowed money from me.’ Morphology4. The languages under study in this paper are not morphologically very rich, and this is reflected in their demonstrative systems. The morphology of the demonstratives is basically invariable. In Ga and Akan, except for the demonstrative determiner and pronoun categories, there are completely different demonstrative forms functioning in different syntactic contexts. Of course, the MDAs and the ‘particular’ demonstratives are identical in Akan and Ga, and nearly identical in Ewe. Thus, the various categories do not present as inflections of a basic form. In Akan, the demonstrative pronouns have the same stem as the determiners, and in addition, they have prefixes consistent with the nominal prefix system in the language. A derivation of the Ga demonstrative determiners from the pronouns or vice versa is not so transparent, even though there is some phonological similarity between the forms. The situation is however different in Ewe. The demonstrative determiners appear to form the stem of the pronouns as well as the LDAs. All the proximal forms include the morpheme sia, and the distals have ma. The demonstrative identifier is also quite similar to the proximal demonstrative pronoun, except that it stem has a single vowel instead of two found in the other proximal demonstratives. It is the MDA which is significantly different in form from the other categories of demonstratives. Table 1 provides a summary of the categories of demonstratives in Akan, Ga and Ewe. 200 Legon Journal of the HUMANITIES Volume 20 (2009) Deter- miners Pron- ouns LDAs MDAs Parti- cular Iden- tifier Akan yi, no oyi/eyi/ wei/iyi, ɔno/εno εha, εhɔ dei,dεm (FA) sei,saa (AS,AK) dεm/saa ni(e) /nyi Ga nεε, lε εnε/ mεnε, no/lε biε, jεmε nεkε, nakai nεkε, nakai nε Ewe sia, ma esia, ema afisia, afima, afimε ale, sigbe/ nenema nenem esi Table 1: Categories of Demonstratives Semantics5. According to Diessel (1999), the semantic features embedded in demonstratives are basically of two kinds: deictic and qualitative features. The deictic feature inherent in demonstratives is what enables a speaker who uses a demonstrative form to indicate the relative distance of an object or location in relation to a deictic centre (origo). The deictic centre often tends to be closely related to the location of the speaker, even though more elaborate demonstrative systems may locate objects and locations in relation to the addressee as well. The demonstrative systems in Akan, Ga and Ewe basically indicate a two-way deictic contrast, even though Ewe has a three-way contrast for its LDAs. The demonstrative pronouns and determiners all show a two-way contrast; there is a proximal member and what may be considered as its distal counterpart in each set. With regard to the LDAs, Akan and Ga indicate a two-way contrast, whereas Ewe makes a three-way distinction: close to the speaker, away from the speaker, and further away from both speaker and addressee. Akan, Ga and Ewe have deictically distinct MDAs. The distinction is based on whether it is the speaker or someone other than her who is doing the demonstration of the action that accompanies the MDAs in their exophoric Nana Aba Appiah Amfo and Stella Boateng Norgbe 201 (gestural) usage. Ewe and Akan, unlike Ga, does not carry on the deictic contrast associated with the MDAs on to the ‘particular’ demonstrative. As a result, the Akan and Ewe ‘particular’ demonstratives are not deictically contrastive, but Ga has proximal and distal ‘particular’ demonstratives which collocate with their respective demonstrative determiner counterparts. The demonstrative identifiers in all three languages are deictically neutral. This may be due to the fact that they are used to draw the addressee’s attention to objects which are present in the extra-linguistic context, and these are often physically relatively close to the interactants. The deictic information included in the demonstratives is summarised in the table 2 below. Determiners Pronouns LDA MDA ‘Particular’ Identifier Akan 2-way 2-way 2-way 2-way NO NO Ga 2-way 2-way 2-way 2-way 2-way NO Ewe 2-way 2-way 3-way 2-way NO NO Table 2: Deictic Contrast of Demonstratives According to Diessel, the qualitative features inherent in demonstratives help in providing classificatory information about the referent of that demonstrative (phrase). The qualitative features he lists include the following: ontology, animacy, humanness, sex, number and boundedness. Demonstratives in the Kwa languages under study here do not abound with such qualitative features. The features which are most prominent in these demonstratives are those of humanness, and to a lesser extent number. The demonstrative pronouns in Akan and Ga are distinguished on the basis of humanness, even though there are some dialectal peculiarities when it comes to Akan. Akuapem uses oyi ‘this person’ for human entities, and it reserves eyi ‘this one’ for non-human entities. However the Asante and Fante equivalents, wei and iyi respectively, are used in reference to all entities irrespective of their humanness status. With regards to the distal demonstrative pronouns, Asante and Akuapem employ ɔno ‘that person’ for human entities and εno ‘that one’ is used 202 Legon Journal of the HUMANITIES Volume 20 (2009) when the referents are non-human. Here again, Fante makes no such distinction, and ɔno is the appropriate pronoun for both human and non-human entities. The Ga demonstrative pronouns mεnε ‘this person’ and lε ‘that person’ are used in exclusive reference to human entities, while εnε ‘this one’ and ɔno ‘that one’ is employed for non-human entities. As indicated in section 3.2, the Ewe pronouns esia and ema are used in specific reference to non-human entities, thus the Ewe speaker resorts to the use of a demonstrative determiner and the generic noun ame when she intends a human referent. With regard to number, there is hardly any number distinction reflected by the demonstratives. It is only the Fante PDD and PDP that inflects for number, by the addition of the suffix –nom14 to the demonstrative (i)yi, for plural referents. Ewe has a distinct noun phrase structure, from that of Akan and Ga, where it is the final element in the noun phrase that carries that plural marker –wo. Thus, if it so happens that the demonstrative determiner is the last element in the noun phrase, the plural marker will be attached to it. Pragmatics6. Demonstratives are crucial pragmatic indicators in communicative situations. As Diessel (1999: 93) puts it, their primary purpose is to “orient the hearer in the speech situation, focusing his or her attention on objects, locations, or persons”. Demonstratives, Diessel suggests, may be used exophorically or endophorically. In their exophoric use, demonstratives are used to refer to entities in the speech situation, i.e. the extra-linguistic context, and they are significantly accompanied by pointing gestures. Subsumed under the endophoric use are the anaphoric, discourse deictic and recognitional uses. When demonstratives are used anaphorically, the purpose is to “track participants of the preceding discourse” (Diessel 1999: 96). Demonstratives used discourse deictically refer to whole propositions which are part of the larger discourse. They may be anaphoric or cataphoric, in that, the proposition they refer to has been mentioned earlier (in the case of the former), or the speaker is just about to introduce it (in the case of the latter). Significantly, in its cataphoric use, the proposition which the speaker refers to by the demonstrative should be produced by the same speaker, but this need not be the case in its anaphoric use. The recognitional use of demonstratives is relatively less known and studied.15 Demonstratives used recognitionally do not have their referents in the immediate Nana Aba Appiah Amfo and Stella Boateng Norgbe 203 linguistic or extra-linguistic context. Their purpose is to activate specific shared information between speaker and addressee. This specific shared information is not information which is part of the general shared knowledge of the whole speech community. Since the referents of recognitional demonstratives are not ‘activated’, they may occur with relative clauses or other noun modifiers which aid in the easy identification of the intended referent by the addressee. Indeed these demonstratives, Diessel suggests, are adnominal. In Akan, Ewe and Ga demonstrative identifiers are used exophorically. Indeed, this is a primary feature of demonstrative identifiers. Diessel (1999: 84) says that they are “genuine deictic expressions”. Their basic exophoric function is reflected in the various labels assigned to them by researchers: Schuh (1977: 7) calls them “deictic predicators”, Carlson (1994: 160) refers to them as “deictic identifier pronouns” and Rehg (1981: 15) uses the label “pointing demonstratives”. Demonstrative pronouns in these three languages may be used exophorically. Often their use is licensed by the physical presence of the entity being referred to, in the immediate extra-linguistic context. This group of demonstratives may be used discourse deictically, in reference to a proposition which has been expressed earlier in the discourse. The proximal demonstrative determiners when used as discourse deictics may be anaphoric or cataphoric. In its cataphoric use, the demonstrative and its co-referent proposition have to be produced by the same speaker. Diessel intimates that manner demonstratives are regularly used as discourse deictics, and this is the case in Akan, Ewe and Ga. They may be used in reference to a previously uttered proposition. In addition, they are often used exophorically, in that they are accompanied by appropriate gestures. Utterance final and focus positions particularly favour the exophoric use in all three languages. The proximal manner demonstratives, sei/dei (Akan), ale (Ewe), nεkε (Ga), are used when the speaker is the one doing the demonstration and the demonstration is performed simultaneously as the utterance or immediately after. If the speaker performs the action after which she produces the utterance involving the MDA, then the appropriate forms are the distal ones saa/dεm (Akan), nenema/sigbe (Ewe) and nakai (Ga). The demonstrative determiners in the languages could serve both anaphoric and cataphoric purposes. 204 Legon Journal of the HUMANITIES Volume 20 (2009) We have not as yet identified clear recognitional uses of demonstratives in these Kwa languages. However, the strategy where speakers modify a proper name with the distal demonstrative/definite article in Akan and Ga appears to serve a recognitional function. In the following Akan example (53), and its Ga equivalent, in (54), Akua need not have been mentioned in the preceding discourse, neither is her physical presence required to render the utterances felicitous. All that is required is that the referent of Akua is ‘familiar’ to both speaker and addressee. 53) Akua no, ɔ-ba a, o-be-hu. Akua DEF, she-come CM, she-FUT-see ‘That Akua, she will see when she comes.’ 54) Akua lε, kεji e-baa, e-baa-na Akua DDD, CM she-come, she-FUT-see In the following sub-section, we turn our attention to what we believe is the grammaticalization of the Ewe relative clause marker from the distal demonstrative determiner. Grammaticalisation6.1 Grammaticalisation is defined by Meillet (1912: 131) as “the attribution of grammatical character to a previously autonomous word”. It is the process whereby lexical categories become grammatical categories. As Hopper and Traugott’s (2003: 1) recent definition demonstrates, grammaticalisation also involves “how grammatical items develop new grammatical functions”. The effects of grammaticalisation are manifest at the phonological, morphological, syntactic and semantic levels of analysis of a linguistic form (see Heine and Reh 1984, Lehmann 1995, Hopper and Traugott 2003, Sweetser 1988, Heine, Claudi and Hunnemeyer 1991, among others). Diessel attests to various categories of demonstratives grammaticalising into other grammatical categories such as third person pronouns, relative pronouns, complementisers and sentence connectives. We have already hinted at the phonological identity of the Akan and Ga distal demonstrative determiners and the definite articles. Indeed third person object pronouns in these two languages have similar segments as the demonstrative determiners/definite articles, except Nana Aba Appiah Amfo and Stella Boateng Norgbe 205 that the former has low tones, whereas the latter has high tones. We will not delve into the relationship between these forms in this paper, for lack of sufficient data at this point (but see Amfo 2006 for the discussion on the Akan forms). Our focus in this section of the paper is on the grammaticalisation of the Ewe relative clause marker. The relative clause marker in Ewe is si, which we suggest has been derived from the proximal demonstrative determiner sia. Our reasons for arguing this way are three-fold. First, the phonological similarity between these two forms is striking. Second, this view is supported by a co-occurrence restriction on the two forms in the same extended noun phrase. Third, our argument is strengthened by the cross-linguistic tendency for relative pronouns (or markers) to derive from adnominal demonstratives. We will take these in turn. As illustrated in example (22), repeated here for convenience as (55), the referent of a noun modified by sia is activated, either because it has just been mentioned in the immediately preceding utterance or it is in the immediate extra linguistic context and it is relatively close to the speaker. 55) Agbale sia nyo. Book PDD be.good ‘This book is good.’ The relative clause marker, as we have mentioned, is si. Example (56) involves a head noun modified by a determiner ɖevi ma ‘that child’, which is in turn modified by a relative clause introduced by si. Ð56) evi ma si xɔ nɔ∫e gbãtɔ la va zu e-xɔlŸ. Child DDD REL collect position first DCM come become his-friend ‘That child, who was first, has become his friend.’ Certainly, the phonological identity between the proximal demonstrative determiner and the relative clause marker is obvious. What makes our suspicion of a historical relation between these two markers stronger is the fact that the PDD cannot modify a noun which is in turn modified by a relative clause introduced by si. The DDD ma is the only demonstrative that can be used in such relative 206 Legon Journal of the HUMANITIES Volume 20 (2009) constructions. As a result, an utterance like the one in (57) where both sia and si co-occur is ungrammatical. So whether the head noun is relatively close or distant to the speaker is something which the addressee is left to infer. 57)* Ðevi sia si ∫e ŋku-wo gba la-e xɔ Child PDD REL POSS eye-PL break DCM-FM collect nɔ∫e gbãtɔ. position first ‘This child who is blind was first.’ In the other languages under investigation here, and in other languages that we are familiar with, nouns modified by relative clauses can be modified by both proximal and distal demonstrative determiners. The relative clause marker in Akan is a, which is clearly phonologically different from the demonstrative determiners yi and no. In Ga, the DDD is lε, the PDD is nεε and the relative clause marker is ni. The similarity displayed in Ewe between the PDD and the relative clause marker is unparalleled in the other languages. Considering that there is no general restriction on the co-occurrence of PDDs and relative clause markers in these closely related languages, the restriction that pertains in Ewe is in all likelihood due to the close historical relation between the two forms. The derivation of relative clause markers16 from demonstratives is nothing new. Diessel reports (following Lehmann 1984) that the German relative pronoun der derives from a demonstrative pronoun. The suggested pathway from the demonstrative pronoun to a relative clause marker is via an adnominal demonstrative. Diessel concludes then that if Lehmann’s analysis is to be taken seriously then adnominal demonstratives ought to be considered as one source of relative clause markers. This is what we suggest has happened in Ewe, that the PDD is the source of the relative clause marker.17 Nana Aba Appiah Amfo and Stella Boateng Norgbe 207 Conclusion7. This paper has taken a comparative look at the demonstrative systems in three Kwa languages – Akan, Ewe and Ga. The approach has been mostly descriptive, covering the categories of demonstratives that the languages have and the various syntactic positions associated with them. The morphology of the demonstrative systems was considered, as well as certain semantic and pragmatic features of the demonstratives. We conclude that the demonstratives in these languages are generally very similar, even though we observe some interesting differences. Except for the demonstrative identifier category, the demonstrative categories, in most cases, manifest a basic two-way deictic contrast. The exceptions are as follows: the Akan and Ewe ‘particular’ demonstratives are deictically neutral, and the Ewe LDAs display a three-way deictic contrast. It is only Ewe which has basic demonstrative morphemes (distal and proximal) in the determiner, pronoun and locational adverb categories. Akan has closely related demonstrative morphemes in the determiner and pronoun categories only, while Ga does not appear to have a consistent system in this regard. This relative morphological invariability is in consonance with the morphological paucity associated with these languages. The least common of the demonstrative categories – the demonstrative identifier – is found to occur in all the languages. It is an indication that Diessel’s (1999) categorisation (as compared to others, such as Dixon’s (2003)) may be more suitable in describing the demonstrative systems in languages within the Kwa family and other related language families. However, it still falls short of capturing the ‘particular’ demonstrative category, which is present in all three languages, and indications are that it is a common category in languages within the sub-region. The description of the demonstrative systems of these languages could form a basis for further work and an eventual re-construction of the demonstrative system in the proto-language. 208 Legon Journal of the HUMANITIES Volume 20 (2009) NOTES We duly acknowledge the assistance of our language consultants Seth Allotey and 1. Nii Ayi Ayiteh. Amfo (2007) is an e2. xception. Even tho3. ugh there is some variety of Akan spoken in the Ivory Coast, this paper is concerned with Akan as a Ghanaian language. Ewe is widely s4. poken in Southern Togo and Benin as well. Once again, we concentrate on Ewe spoken in Ghana. The dialects which a5. re collectively called Ewe may be put into three main groups: the Southern, Central and Northern dialects. Alternatively, the Central and Northern dialects are considered as the Inland dialects, and the Southern dialects are the Coastal dialects. However, its role as a 6. lingua franca for residents in the capital is gradually ceding to Akan. The abbreviations used in this p7. aper is as follows: CM Conditional marker; COMPL Completive aspect; CONJ Coordinating connective; DCM Dependent clause marker; DDD Distal demonstrative determiner; DDP Distal demonstrative pronoun; DEF Definite article; DI Demonstrative identifier; FM Focus marker; FUT Future tense; HAB Habitual marker; MDA Manner demonstrative adverb; NEG Negation morpheme; PD Particular demonstrative; PDD Proximal demonstrative determiner; POSS Possessive marker; PL Plural; PRO Pronoun; PST Past tense; PROG Progressive marker; RED Reduplication; REL Relative clause marker; TM Topic marker; PERF Perfect aspect; PDP Proximal demonstrative pronoun. For a fuller range8. of the kind of clauses that the demonstrative determiners occur in, see Amfo (2006). In Amfo (2007), the demon9. strative pronouns were distinguished based on animacy features, we believe that the humanness distinction is more specific and accurate. Again, the dialectal variation we mention here was not recognized in the earlier study. This is a deviation from and10. an improvement of what was originally reported in Amfo (2007). Afii is sometime11. s used instead of afisia. Nenema and sigbe a12. re dialectal variants. The former is from the Anlo dialect and the latter is from the most of the inland dialects such as Peki, Kpando and Ho. Nana Aba Appiah Amfo and Stella Boateng Norgbe 209 Ga has another marker, naa, which occurs in non-verbal clauses. This marker appears 13. to function in an identical manner as nE, except that it occurs pre-nominally in contrast to nE which occurs post-nominally. -nom is the p14. lural suffix for kinship nouns in Akan. Some studies15. which make explicit or implicit reference to this use include Gundel et al (1993), Auer (1984) and Lakoff (1974). As Keenan (1985) observe16. s, relative markers may come in the form of pronouns (as is the case in most European languages) or particles. The relative marker in Ewe is an invariable particle. Even though most of the reported studies of grammaticalization relates to pronouns, we believe those principles are still applicable here, even if we are dealing with a particle rather than a pronoun. Unfortunately, the 17. language lacks adequate historical data, and as a result we have to resort to extrapolation. 210 Legon Journal of the HUMANITIES Volume 20 (2009) REFERENCES Amfo, Nana Aba Appiah 2006. Syntactic Variation Versus Semantic Uniqueness: a Study of the Multifunctional Marker no in Akan. In Eva Thue Vold, Gunn Inger Lyse and Anje Muller Gjesdal (eds.). New Voices in Linguistics. New Castle upon Tyne: Cambridge Scholars Publishing. 211-224 Amfo, Nana Aba Appiah 2007. Akan Demonstratives. In Doris L. Payne and Jaime Pen)a (eds.). Selected Proceedings of the 37th Annual Conference on African Linguistics. Sommerville MA: Cascadilla Proceedings Press, 134-148. Anderson, Stephen R. and Edward L. Keenan 1985. Deixis. In Timothy Shopen (ed.) Language Typology and Syntactic Description, vol. III, Grammatical Categories and the Lexicon. Cambridge: Cambridge University Press, 259-308. Auer, Josef. C. P. 1984. Referential Problems in Conversation. Journal of Pragmatics 8, 627-648. Carlson, Robert 1994. A Grammar of Supyire. Berlin: Mouton de Gruyter. Dakubu, Mary Esther K. 1988. Ga-Dangme. In Mary Esther K. Dakubu (ed.) Languages in Ghana, chapter 5. London: Kegan Paul International, 102- 118. Diessel, Holger 1999. Demonstratives: Form, Function and Grammaticalization. John Benjamins Publishing Company. Dixon, Robert M. W. 2003. Demonstratives: A Cross-Linguistic Typology. Studies in language 27:1, 61-112. Duthie, Alan. S. 1988. Ewe. In Mary Esther K. Dakubu (ed.) Languages in Ghana, chapter 4. London: Kegan Paul International, 91-96. Gundel, Jeanette, K., Hedberg, Nancy and Ron Zacharski. 1993. Cognitive Status and the Form of Referring Expressions In Discourse. Language 69:2, 274-307. Heine, Bernd, Ulrike Claudi and Friederike Hünnemeyer. 1991. From cognition to grammar: evidence from African languages. In Elizabeth C. Traugott and Bernd Heine (eds.) Approaches to grammaticalization. Vol. 1, 149- Nana Aba Appiah Amfo and Stella Boateng Norgbe 211 188. [Typological Studies in Language 19]. Amsterdan: John Benjamins. Heine, Bernd and Mechthild Reh 1984. Grammaticalization and reanalysis in African languages. Hamburg: Helmut Buske Verlag. Hopper, Paul J and Elizabeth C. Traugott 2003. Grammaticalization. Cambridge: Cambridge University Press. Jaggar, Philip J. 2001. Hausa. London Oriental and African Language Library 7 Amsterdam: John Benjamins. Keenan, Edward L. 1985. Relative Clauses. Language Typology and Syntactic Description, vol. II, Cambridge: Cambridge University Press, 147-170. Lakoff, Robin 1974. Remarks on this and that. Chicago Linguistic Society 10, 345-356. Lehmann, Christian 1984. Der Relativsatz: Typologie seiner Strkturen, Theorie seiner Funktionen, Kompendium seiner Grammatik. Tübingen: Narr. Lehmann, Christian 1995. Thoughts on Grammaticalization. Munich: Lincom Europa. Lyons, John 1977. Semantics, vol. I & II. Cambridge: Cambridge University Press. Meillet, Antoine 1912. L’évolution des formes grammaticales. In Antoine Meillet, Linguistiques historique et linguistique général, 130-148. Paris : Champion. Obeng, Samuel Gyasi 2005. Akan and Nyo languages. In Philipp Strazny (ed.) Encyclopedia of Linguistics, 28-31. New York: Routledge. Rehg, Kenneth, L. 1981. Ponapean Reference Grammar. Honolulu: University of Hawaii Press. Schuh, Russel. G. 1977. Bade/Ngizim determiner system. Afroasiatic Linguistics 4-3, 1-74. Stewart, John M. 1989. Kwa. In John Bendor-Samuel (ed.) The Niger-Congo languages: a classification and description of Africa’s largest family, 216-245. Lanham: University Press of America. Sweetser, Eve E. 1988. Grammaticalization and Semantic Bleaching. Berkeley Linguistics Society, 14:389-405